From tjrc at sanger.ac.uk Tue Apr 1 02:35:31 2008 From: tjrc at sanger.ac.uk (Tim Cutts) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] "Code" vs. "Codes" In-Reply-To: References: <47F14B5E.7010904@berkeley.edu> <6.2.3.4.2.20080331153851.0289a130@mail.jpl.nasa.gov> <47F174CA.3070304@berkeley.edu> <20080401002824.GA15924@bx9.net> Message-ID: <306456F7-E9AD-4D7E-B0D4-F3D492BD5A59@sanger.ac.uk> OT: There's an interesting article on the evolution of English in this week's New Scientist, which addresses some of these very observations. Their contention is that because English is now a global language with a very large number of non-native speakers, it's regularising quite rapidly, and the appearance of "regular" plurals like "e-mails" is just such an example. Apparently words like "sugar" are called "mass nouns" -- as opposed to countable nouns -- and it's quite an awkward concept even for native English speakers to get right. How often do you hear someone say "less people" rather than the strictly correct "fewer people"?. I suspect the distinction is gradually disappearing from the language. The article goes on to suggest that one possible evolutionary path English might follow is similar to what happened to Latin in the early middle ages. Once the major political power using Latin had disappeared, the language rapidly fragmented into today's romance languages. The same could easily happen to English, once the global dominant power shifts to asia, which many people seem to think is inevitable. Anyway, back to Beowulf... Tim -- The Wellcome Trust Sanger Institute is operated by Genome Research Limited, a charity registered in England with number 1021457 and a company registered in England with number 2742969, whose registered office is 215 Euston Road, London, NW1 2BE. From gerry.creager at tamu.edu Tue Apr 1 05:40:35 2008 From: gerry.creager at tamu.edu (Gerry Creager) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] "Code" vs. "Codes" In-Reply-To: <47F174CA.3070304@berkeley.edu> References: <47F14B5E.7010904@berkeley.edu> <6.2.3.4.2.20080331153851.0289a130@mail.jpl.nasa.gov> <47F174CA.3070304@berkeley.edu> Message-ID: <47F22D43.5090608@tamu.edu> Jon Forrest wrote: > I appreciate everyone's comments so far, but please > keep in mind that I only meant to comment on the > use of "codes" (plural). The use of "code" in place > of "program" feels normal. > > I still suspect that somewhere this started with > a non-native speaker of English. The same thing > has happened recently with "e-mail", which didn't > used to exist in common usage in plural form. > Now you see "e-mails" used by all kinds of people. I started hearing 'codes' when I became more active in the modeling community. I've heard the term 'code' as synonymous to 'program' since, well, I started learning to 'code' in about 1966 or 1967 (yes, in IBM FORTRAN; I'm that old). When I wrote my first weather model, there was no one to chat about it with save other students, and they weren't a lively bunch. When I wrote another model, for a specific application, however, I began talking to some folks at Rice University about CFD and they referred to different 'codes' as differentiation among both various programs, and physics approaches. In the weather community I've heard the term used to describe that there are several physics approaches to solution of the various problems. And its use seems to make perfect sense in that context. We've also used it to describe differing approaches (eg., explicit vs. implicit; finite element vs difference, etc). I suspect some folks aren't as precise (I've been known to be sloppy in speach) as some of the folks I've been around, in regard to this. gerry -- Gerry Creager -- gerry.creager@tamu.edu Texas Mesonet -- AATLT, Texas A&M University Cell: 979.229.5301 Office: 979.458.4020 FAX: 979.862.3983 Office: 1700 Research Parkway Ste 160, TAMU, College Station, TX 77843 From gerry.creager at tamu.edu Tue Apr 1 05:46:22 2008 From: gerry.creager at tamu.edu (Gerry Creager) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] "Code" vs. "Codes" In-Reply-To: References: <47F14B5E.7010904@berkeley.edu> <6.2.3.4.2.20080331153851.0289a130@mail.jpl.nasa.gov> <47F174CA.3070304@berkeley.edu> <20080401002824.GA15924@bx9.net> Message-ID: <47F22E9E.1010008@tamu.edu> Greg! You had to launch the clone? WHAT have you done? ;-) gerry Robert G. Brown wrote: > On Mon, 31 Mar 2008, Greg Lindahl wrote: > >> English has super-plurals: >> >> http://d.a.nicolas.free.fr/research/Linnebo-Nicolas-Superplurals.pdf >> >> rgb knows this, but he was pretending like he doesn't use "all-y'all" >> all the time... > > Y'all find some mighty odd things out there on the web, y'know Greg? I > hadn't ever heard of super-plurals -- very interesting. All y'all > should look at this, because the bit on the analysis of certain > super-plural constructs providing a lead to second and third (and > speculatively higher order) logics is mighty interesting. > > Still, God knows that it is difficult enough to keep my verbs all tensed > up and correctly quantified without superplural constructs. Now > Whitehead and Russell, Whitehead and Hilbert, is a superplural construct > in a self-referential sentence that asks whether the verb "is" that > refers to Whitehead and Russell, Whitehead and Hilbert, should correctly > be singular super-super-plural or if I should have used "are" to > indicate ordinary super-plural (and while I'm at it, whether or not I > can use "used" as a self-referential past tense of a usage in a single > sentence with an implicit temporal order of expression right before > switching back to present tense). And only Kilpatrick Knows if I should > have put a question mark at the end...;-) > > rgb > >> Anyway, the reason I suggested the OED is that they often have early >> usage quotes, which may well include examples of 'codes'. I can only >> access the free version, which shows that they do have the definition, >> but doesn't include the examples. > > I have VPN problems at home these days (too complicated to describe, > sigh) but when I get to Duke tomorrow I'll see if the University > subscribes to the OED so I can use it for free. > > rgb > >> >> -- greg >> _______________________________________________ >> Beowulf mailing list, Beowulf@beowulf.org >> To change your subscription (digest mode or unsubscribe) visit >> http://www.beowulf.org/mailman/listinfo/beowulf >> > -- Gerry Creager -- gerry.creager@tamu.edu Texas Mesonet -- AATLT, Texas A&M University Cell: 979.229.5301 Office: 979.458.4020 FAX: 979.862.3983 Office: 1700 Research Parkway Ste 160, TAMU, College Station, TX 77843 From henning.fehrmann at aei.mpg.de Tue Apr 1 07:18:16 2008 From: henning.fehrmann at aei.mpg.de (Henning Fehrmann) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] SNMPtraps Message-ID: <20080401141816.GA4045@gretchen.aei.uni-hannover.de> Hi all, A server collecting the SNMPtraps are usually writing into /var/log/syslog or other log files. Some times the messages demand an urgent reaction. Are there tools which handle incoming traps immediately by starting scripts to , e.g, send out an email or a SMS? Cheers Henning From amacater at galactic.demon.co.uk Tue Apr 1 12:23:18 2008 From: amacater at galactic.demon.co.uk (Andrew M.A. Cater) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] "Code" vs. "Codes" In-Reply-To: <873aq62mvt.fsf@nyarlothotep.fyndo.com> References: <47F14B5E.7010904@berkeley.edu> <6.2.3.4.2.20080331153851.0289a130@mail.jpl.nasa.gov> <47F174CA.3070304@berkeley.edu> <873aq62mvt.fsf@nyarlothotep.fyndo.com> Message-ID: <20080401192318.GA2825@galactic.demon.co.uk> On Tue, Apr 01, 2008 at 12:57:10AM -0400, Eric Moore wrote: > Jon Forrest writes: > > > I found this: > > This German Flag Officer is to be accompanied by a Communications > Officer who is familiar with the German Naval W/T organization and > who is to bring with him the current naval communications Orders, > including allocation of frequencies, list of W/T and R/T call signs > in force, and a list of all codes and cyphers in use, and intended > to be brought into use. > > So certainly the use of "codes" as a plural for "code" was preexisting > in the English language by the time "code" became a synonym for > "program", so if "code" for "program" is accepted, there's no reason > to look to non-native speakers for "codes" for "programs". > Yes, but this is "code" as in "thing to keep other things secret" cryptography, cryptanalysis, "code breaking" "code" as in "I'm finding it hard to code this thing up / the code is all in asm rather than C" is standard computer speak in the US and English speaking world. "codes" as in "the weather forecasting / CFD / higher order topology computer programs that I've been working on" - feels utterly alien but then "code" and "codex" are cognate. Compact OED 1996 gives ( [] surround italicised text in the original, a. and b. are bold text a and b in the original ) code [n] 1 a system of words, letters, figures, or symbols used to represent others for secrecy or brevity 2 a system of pre-arranged signals esp. used to ensure secrecy in transmitting messages 3 [Computing] a piece of program text 4 a systematic collection of laws etc. 5 a. the prevailing morality of a society or class (code of honour) b. a person's standard of moral behaviour [v.tr] put (a message, program etc.) into code (from Latin [codex] coder [n] Where's a tame linguistics professor when you need him - paging Martin Wheeler, paging Martin Wheeler :) Andy From lindahl at pbm.com Tue Apr 1 13:21:54 2008 From: lindahl at pbm.com (Greg Lindahl) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] "Code" vs. "Codes" In-Reply-To: <306456F7-E9AD-4D7E-B0D4-F3D492BD5A59@sanger.ac.uk> References: <47F14B5E.7010904@berkeley.edu> <6.2.3.4.2.20080331153851.0289a130@mail.jpl.nasa.gov> <47F174CA.3070304@berkeley.edu> <20080401002824.GA15924@bx9.net> <306456F7-E9AD-4D7E-B0D4-F3D492BD5A59@sanger.ac.uk> Message-ID: <20080401202154.GA5513@bx9.net> On Tue, Apr 01, 2008 at 10:35:31AM +0100, Tim Cutts wrote: > and the appearance of "regular" plurals > like "e-mails" is just such an example. I think emails is an exception. For postal mail, you'd say "I received 3 letters". For email, we don't call them letters, we just use "emails." But I still say "I deleted all my email." It's not becoming more regular, not unless I start saying weird things like "emails server" and "emails database". > Once the major political power using Latin had > disappeared, The Church was still a major political power using Latin, not to mention the Holy Roman Empire and the Byzantine Empire -- and Latin itself evolved steadily. -- g From mwheeler at startext.co.uk Tue Apr 1 14:15:35 2008 From: mwheeler at startext.co.uk (Martin Wheeler) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] "Code" vs. "Codes" In-Reply-To: <20080401192318.GA2825@galactic.demon.co.uk> References: <47F14B5E.7010904@berkeley.edu> <6.2.3.4.2.20080331153851.0289a130@mail.jpl.nasa.gov> <47F174CA.3070304@berkeley.edu> <873aq62mvt.fsf@nyarlothotep.fyndo.com> <20080401192318.GA2825@galactic.demon.co.uk> Message-ID: On Tue, 1 Apr 2008, Andrew M.A. Cater wrote: > Where's a tame linguistics professor when you need him - paging > Martin Wheeler, paging Martin Wheeler :) This "tame linguistics professor" is currently fixing the recalcitrant codes of a local website, when in fact he would rather be sat in an armchair reading a good book. Or painting. (And btw, there's a *lot* more to linguistics than historical etymology, as well you should know.) But funnily enough, it was on this very list that I first noticed the use of both 'code' and 'codes' as almost a synonym for what I would usually term 'software' or 'program(s)' myself. As I'm no specialist in the field, I noted the usage, but said nothing, and asked no questions. (Descriptive; not prescriptive.) I also made a mental note that it was probably an Americanism. (Always a good device to cover up one's own ignorance.) But I'm willing to bet it has nothing at all to do with native-speakers or non-native-speakers. For starters, it's one of those lexical items that you find only in use among specialist users of the language; and specialists are almost always extremely precise in their usage. (Too precise. If two similar terms exist to describe the same thing, they'll start differentiating between them. Cf. the usage of 'I learned' vs 'I learnt' amongst pedantic professors of English. Non-native-speakers [in general] don't care.) Then there's the fact that there are all sorts of native-speakers. I'm sure that the local yokels of N. Carolina (US) would be completely non-plussed by the speech of local yokels of Somerset (UK) -- heck, as a northerner myself, I still get baffled at times. Yet we're all native-speakers. Linguistically, we share a common core -- nevertheless, we all have our own particular usages in certain areas. The danger is always that one is tempted to take one's own idiolect for the national norm, and be completely taken by surprise when encountering any sort of neologism. (And I'd be greatly surprised if it were any sort of attempt to resolve the 'program/programme' dichotomy -- that only occurs in written English, anyway.) Regards, -- Martin Wheeler - G5FM +44 1458 83-1103 - Glastonbury - BA6 9PH - England mwheeler@startext.co.uk http://martinwheeler.net/ http://avalonit.net/ GPG pub key : 01269BEB 6CAD BFFB DB11 653E B1B7 C62B AC93 0ED8 0126 9BEB - Share your knowledge. It's a way of achieving immortality. - From ascheinine at tuffmail.us Tue Apr 1 16:55:05 2008 From: ascheinine at tuffmail.us (Alan Louis Scheinine) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] "Code" vs. "Codes" In-Reply-To: <873aq62mvt.fsf@nyarlothotep.fyndo.com> References: <47F14B5E.7010904@berkeley.edu> <6.2.3.4.2.20080331153851.0289a130@mail.jpl.nasa.gov> <47F174CA.3070304@berkeley.edu> <873aq62mvt.fsf@nyarlothotep.fyndo.com> Message-ID: <47F2CB59.1000601@tuffmail.us> I can't stand it any longer, I've got to make my "contribution" to this discussion. Eric Moore cites the text > a list of all codes and cyphers in use For this reason, I always thought that using "code" for programs was considered sloppy jargon and that in a formal document or article one should write "computer program" rather than "code". -- Alan Centro di Ricerca, Sviluppo e Studi Superiori in Sardegna Center for Advanced Studies, Research, and Development in Sardinia Alan Scheinine c/o CRS4 Loc. Pixina Manna Edificio 1 09010 Pula (Cagliari), Italy Email: ascheinine@acm.org Phone: 070 9250 238 [+39 070 9250 238] Fax: 070 9250 216 or 220 [+39 070 9250 216 or +39 070 9250 220] Operator at reception: 070 9250 1 [+39 070 9250 1] Mobile phone: 347 7990472 [+39 347 7990472] From amacater at galactic.demon.co.uk Tue Apr 1 16:11:42 2008 From: amacater at galactic.demon.co.uk (Andrew M.A. Cater) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] "Code" vs. "Codes" In-Reply-To: <47F2CB59.1000601@tuffmail.us> References: <47F14B5E.7010904@berkeley.edu> <6.2.3.4.2.20080331153851.0289a130@mail.jpl.nasa.gov> <47F174CA.3070304@berkeley.edu> <873aq62mvt.fsf@nyarlothotep.fyndo.com> <47F2CB59.1000601@tuffmail.us> Message-ID: <20080401231142.GA7306@galactic.demon.co.uk> On Wed, Apr 02, 2008 at 01:55:05AM +0200, Alan Louis Scheinine wrote: > I can't stand it any longer, I've got to make > my "contribution" to this discussion. > Eric Moore cites the text > > a list of all codes and cyphers in use > For this reason, I always thought that > using "code" for programs was considered sloppy jargon > and that in a formal document or article one should > write "computer program" rather than "code". > As Martin Wheeler says, you also have to consider the common understanding and specialist vocabulary of your specific audience: "Code re-use" - may make a computer scientist excited/ill at the thought, depending on his viewpoint of the desirability of this. It may have the same effects on cryptanalysts and computer security consultants - but for entirely different reasons :) Andy - once a trainee lawyer - which profession has been reusing portions of Justinian's Code for millennia. Andy From xclski at yahoo.com Tue Apr 1 19:08:40 2008 From: xclski at yahoo.com (Ellis Wilson) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] HPL Benchmarking and Optimization Message-ID: <456932.47911.qm@web37904.mail.mud.yahoo.com> I'm currently running some benchmarks on my cluster, one of special interest to me being High Performance Linpack, implemented for parallel installations. I've gotten a hang of the variables (Ns, NBs, etc.) from information available online, however would be really interested to hear back on whether my results are far under par and if compilation optimizations (or different fortran compilers) will increase my performance vastly. Currently I get these kind of numbers from tested computers using the same environment (gentoo, fortran in gcc, hpl, all same compilation options): 1 x Core2Duo (2.1ghz/core, 2gigs ram) - 2.3Gflops 1 x Athlon 64 3500+ (2.2ghz, 1gig ram) - 1.0Glops 4 x Core2Duo (2.1ghz/core for a total of 8 cores, 2gigs ram/node, 100mbit Ethernet interconnect) - 6.7Gflops I'm finding numbers online that are well below, at, and well above my own. Are these numbers low and if so, how could I improve the situation (or should I attempt to optimize at all, this seems a bit backwards since it's making the benchmark akin to my application)? Thanks in advance, Ellis ____________________________________________________________________________________ You rock. That's why Blockbuster's offering you one month of Blockbuster Total Access, No Cost. http://tc.deals.yahoo.com/tc/blockbuster/text5.com From gerry.creager at tamu.edu Wed Apr 2 06:33:34 2008 From: gerry.creager at tamu.edu (Gerry Creager) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] "Code" vs. "Codes" In-Reply-To: <47F2CB59.1000601@tuffmail.us> References: <47F14B5E.7010904@berkeley.edu> <6.2.3.4.2.20080331153851.0289a130@mail.jpl.nasa.gov> <47F174CA.3070304@berkeley.edu> <873aq62mvt.fsf@nyarlothotep.fyndo.com> <47F2CB59.1000601@tuffmail.us> Message-ID: <47F38B2E.1050600@tamu.edu> Hardly sloppy use, 'code' as a reference to 'program' (n.) may be a very precise reference to how some people write programs. Some of the 'code' I see is, in fact, cryptic to the point of obfuscation. gerry Alan Louis Scheinine wrote: > I can't stand it any longer, I've got to make > my "contribution" to this discussion. > Eric Moore cites the text > > a list of all codes and cyphers in use > For this reason, I always thought that > using "code" for programs was considered sloppy jargon > and that in a formal document or article one should > write "computer program" rather than "code". > > -- Alan > > Centro di Ricerca, Sviluppo e Studi Superiori in Sardegna > Center for Advanced Studies, Research, and Development in Sardinia > > Alan Scheinine > c/o CRS4 > Loc. Pixina Manna Edificio 1 > 09010 Pula (Cagliari), Italy > > Email: ascheinine@acm.org > > Phone: 070 9250 238 [+39 070 9250 238] > Fax: 070 9250 216 or 220 [+39 070 9250 216 or +39 070 9250 220] > Operator at reception: 070 9250 1 [+39 070 9250 1] > Mobile phone: 347 7990472 [+39 347 7990472] > > _______________________________________________ > Beowulf mailing list, Beowulf@beowulf.org > To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf -- Gerry Creager -- gerry.creager@tamu.edu Texas Mesonet -- AATLT, Texas A&M University Cell: 979.229.5301 Office: 979.862.3982 FAX: 979.862.3983 Office: 1700 Research Parkway Ste 160, TAMU, College Station, TX 77843 From kalpana0611 at gmail.com Wed Apr 2 06:35:34 2008 From: kalpana0611 at gmail.com (Cally K) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] SMPs + One processor machines = Heterogeneous Cluster Message-ID: Hi guys, Thanks for the previous answers to my question, did not expect that kinda flow of information and it proved to be helpful in my research. I have a different question and I hope I can get some nice feedback. I am dealing with heterogeneous cluster in my research. I am a novice when it comes to networking, and have only been around with cluster for about a year. Anyway, from my previous question, and after doing some reading, can I say that dual cores and quad cores are known as SMPs.? And would 1 say that, a cluster that has machines with 1 processor ( etc. Pentium 4 )and 2 SMPS( it can be quad or dual ) but all belonging to one family ( intel for example ) -- can I say that the cluster is a heterogeneous cluster. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.scyld.com/pipermail/beowulf/attachments/20080402/128b8035/attachment.html From richard.walsh at comcast.net Wed Apr 2 07:14:17 2008 From: richard.walsh at comcast.net (richard.walsh@comcast.net) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] SMPs + One processor machines = Heterogeneous Cluster Message-ID: <040220081414.19245.47F394B9000A13B700004B2D2212059214089C040E99D20B9D0E080C079D@comcast.net> Skipped content of type multipart/alternative-------------- next part -------------- An embedded message was scrubbed... From: "Cally K" Subject: [Beowulf] SMPs + One processor machines = Heterogeneous Cluster Date: Wed, 2 Apr 2008 13:43:59 +0000 Size: 725 Url: http://www.scyld.com/pipermail/beowulf/attachments/20080402/9ddf88f7/attachment.mht From rachelle at ural.wustl.edu Tue Apr 1 12:46:37 2008 From: rachelle at ural.wustl.edu (Rachelle Hinrichs) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] "Code" vs. "Codes" In-Reply-To: <200804011900.m31J07BX028280@bluewest.scyld.com> References: <200804011900.m31J07BX028280@bluewest.scyld.com> Message-ID: > From: Greg Lindahl > > English has super-plurals: > > http://d.a.nicolas.free.fr/research/Linnebo-Nicolas-Superplurals.pdf > > rgb knows this, but he was pretending like he doesn't use "all-y'all" > all the time... > > Anyway, the reason I suggested the OED is that they often have early > usage quotes, which may well include examples of 'codes'. I can only > access the free version, which shows that they do have the definition, > but doesn't include the examples. > > -- greg > Wow, that English article reads like a math paper... This thread is fascinating! It seems to be a convergence of my two favorite email lists: Beowulf and A Word A Day. Y'all are fantastic! And RGB, you never cease to amaze me. You can get more delicious English language tidbits served up at wordsmith.org/awad/ Rachelle (both a regular computer geek, and a word nerd, and apparently, not alone! ;) -- Rachelle Hinrichs Department of Genetics Washington University School of Medicine, Saint Louis, MO rachelle@ural.wustl.edu From mark.kosmowski at gmail.com Wed Apr 2 09:43:56 2008 From: mark.kosmowski at gmail.com (Mark Kosmowski) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] Re: SMPs + One processor machines = Heterogeneous Cluster Message-ID: > Message: 8 > Date: Wed, 2 Apr 2008 21:35:34 +0800 > From: "Cally K" > Subject: [Beowulf] SMPs + One processor machines = Heterogeneous > Cluster > To: beowulf@beowulf.org > Message-ID: > > Content-Type: text/plain; charset="iso-8859-1" > > Hi guys, > > Thanks for the previous answers to my question, did not expect that kinda > flow of information and it proved to be helpful in my research. I have a > different question and I hope I can get some nice feedback. > > I am dealing with heterogeneous cluster in my research. I am a novice when > it comes to networking, and have only been around with cluster for about a > year. Anyway, from my previous question, and after doing some reading, can I > say that dual cores and quad cores are known as SMPs.? > > And would 1 say that, a cluster that has machines with 1 processor ( etc. > Pentium 4 )and 2 SMPS( it can be quad or dual ) but all belonging to one > family ( intel for example ) -- can I say that the cluster is a > heterogeneous cluster. My limited (and perhaps incorrect) understanding is that SMP denotes more than one processor, regardless the number of cores that the processor has. Heterogenous, to me, implies a cluster of machines running different operating systems and/or different architectures. As an example, my personal is three Opteron systems, each of which has two single core CPU's, all running OpenSUSE 10.2 64-bit. This is a homogenous cluster of SMP machines. I plan to add a dual core, single CPU Athlon (64) system to the cluster at some point. This planned machine would (in my opinion) not be SMP (although the more I think about it, the more I think I am either wrong or in the minority camp about SMP and cores) and the cluster would remain homogenous even if I were to put OpenSUSE 11.0 (64 bit) on the system (the new OpenSUSE release and my projected budget have similar timescales). Even if I were to use an Intel Pentium system of whatever nomenclature that is instruction compatible with my existing Opterons (the difficulty in immediately knowing whether a Pentium chip is 64-bit capable or not is one of the reasons I am sticking with AMD) the system would remain homogenous. If, however, I were to add a Sun SPARC machine as a node, now I have a heterogenous cluster. Likewise, if my new Athlon (64) node used Solaris (for Intel), the cluster would be heterogenous in my opinion. My rule of thumb for heterogeneity is whether the programs to be run need to be recompiled for a given node. If all of the nodes can use one single compile of all of the codes needed, the cluster is not heterogenous. I guess this would mean that running a 32-bit program (I want karma for minding my usage of code vs. program here! :) ) on a mixture of 32-bit and 64-bit machines would be homogenous, but if on the same cluster, the 32-bit machines ran a 32-bit program and the 64-bit machines ran a 64-bit program then that cluster might be considered heterogenous. I hope this at least sparks an interesting conversation, Mark Kosmowski From xclski at yahoo.com Wed Apr 2 12:23:34 2008 From: xclski at yahoo.com (Ellis Wilson) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] HPL Benchmarking and Optimization Message-ID: <851490.13543.qm@web37910.mail.mud.yahoo.com> Ellis Wilson wrote: > Currently I get these kind of numbers from tested > computers using the > same environment (gentoo, fortran in gcc, hpl, all > same compilation > options): > 1 x Core2Duo (2.1ghz/core, 2gigs ram) - 2.3Gflops > 1 x Athlon 64 3500+ (2.2ghz, 1gig ram) - 1.0Glops > 4 x Core2Duo (2.1ghz/core for a total of 8 cores, > 2gigs ram/node, > 100mbit Ethernet interconnect) - 6.7Gflops Sorry to double post all, however, I realized my issue involved running HPL on the reference library of BLAS that is generic for every architecture and didn't want to waste anyones time. Giving Portage the benefit of the doubt, I had failed to check that it's dependencies were best for HPL. Following an install of ATLAS and relinking to its libraries, I've gotten the following numbers: 1 x Athlon64 3500+ (2.2ghz, 1gig ram) - 3.6GFlops 1 x Phenom9600 Quadcore (2.3ghz/core, 2gigs ram) - 11.9GFlops I'll likely try MKL soon for the Intel processors I'm interested in. The phenom9600 had previously only gotten 4.5 GFlops, and when I tested it the second time I simply used the same environment I had compiled for the athlon64. Certainly compiling ATLAS native on the phenom will increase the result, hopefully about 350% like with the athlon64 (though I suspect things will be interesting due to bandwidth, etc for quadcores). Anyway, not to end the thread I still am wondering: Do those of you who have professional installations or even simply large setups that are unsure of the exact code which will be run upon your cluster utilize compilation options such as -O3, funroll-loops, -fomit-frame-pointer, etc? Thanks, Ellis ____________________________________________________________________________________ You rock. That's why Blockbuster's offering you one month of Blockbuster Total Access, No Cost. http://tc.deals.yahoo.com/tc/blockbuster/text5.com From tom.elken at qlogic.com Wed Apr 2 13:30:32 2008 From: tom.elken at qlogic.com (Tom Elken) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] HPL Benchmarking and Optimization In-Reply-To: <851490.13543.qm@web37910.mail.mud.yahoo.com> References: <851490.13543.qm@web37910.mail.mud.yahoo.com> Message-ID: <6DB5B58A8E5AB846A7B3B3BFF1B4315A01E4985A@AVEXCH1.qlogic.org> > -----Original Message----- > From: beowulf-bounces@beowulf.org > [mailto:beowulf-bounces@beowulf.org] On Behalf Of Ellis Wilson > > I'll likely try MKL soon for the Intel processors I'm > interested in. Good idea. You might also want to try "Goto BLAS" (Google that to find the free download site). It can be compiled for a different architecture a lot quicker than ATLAS, and provides very good performance for both Intel and AMD architectures. As you may have already found, once you are using a good BLAS library with HPL, various compilers or compiler options won't make much difference in performance. -Tom From jlforrest at berkeley.edu Wed Apr 2 15:04:28 2008 From: jlforrest at berkeley.edu (Jon Forrest) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? Message-ID: <47F402EC.3050701@berkeley.edu> First of all, I like Microsoft, and I voluntarily use Vista as my desktop of choice. I've built and run the Windows environments for the top CS and Civil Engineering departments in the US, and I was the first to port Postgres to Windows NT. That said, I just don't see how Microsoft's HPC server can succeed. I'm not saying this for technical reasons, as I'm sure Microsoft, with enough work, can build a clustering environment that will work just fine. But, why would anybody buy a Windows cluster when there are so many great clustering environments (e.g. Rocks, Perceus, Unicluster Express, ...) and so many cluster-related packages (*MPI, SGE, PBS, gcc, Torque, ...) available for free? What's more, from what I can see, there is very little non-Microsoft-sponsored development going on in HPC computing. Microsoft recently announced (somewhere, I can't find it) the availability of a test cluster for universities to use for financial applications. I bet they get some interest since many business schools use Windows, plus the cluster is free. But, the question remains. How can Microsoft compete with free? How much better will they have to be than standard Linux clusters before they get any mainstream interest? What technical features could they add that couldn't be added to a Linux cluster? Cordially, -- Jon Forrest Research Computing Support College of Chemistry 173 Tan Hall University of California Berkeley Berkeley, CA 94720-1460 510-643-1032 jlforrest@berkeley.edu From landman at scalableinformatics.com Wed Apr 2 15:24:37 2008 From: landman at scalableinformatics.com (Joe Landman) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <47F402EC.3050701@berkeley.edu> References: <47F402EC.3050701@berkeley.edu> Message-ID: <47F407A5.9090403@scalableinformatics.com> Hi Jon Jon Forrest wrote: > First of all, I like Microsoft, and I voluntarily use > Vista as my desktop of choice. I've built and run the > Windows environments for the top CS and Civil Engineering > departments in the US, and I was the first to port > Postgres to Windows NT. Postgres is nice. Did you do a native (ms C compiler) port or a cygwin port? > That said, I just don't see how Microsoft's HPC server > can succeed. I'm not saying this for technical reasons, > as I'm sure Microsoft, with enough work, can build > a clustering environment that will work just fine. > But, why would anybody buy a Windows cluster when > there are so many great clustering environments (e.g. > Rocks, Perceus, Unicluster Express, ...) and so many > cluster-related packages (*MPI, SGE, PBS, gcc, > Torque, ...) available for free? What's more, from > what I can see, there is very little non-Microsoft-sponsored > development going on in HPC computing. Oddly enough, for a long time I have believed (and we have proposed privately to Microsoft) ways in which I think it could make a difference. In short, there are codes that are now and will always be windows based, and that is fine. They may need to run on a cluster, and they should be able to. > Microsoft recently announced (somewhere, I can't find it) > the availability of a test cluster for universities to > use for financial applications. I bet they get some > interest since many business schools use Windows, plus > the cluster is free. > > But, the question remains. How can Microsoft compete with free? It needs to adapt. But not necessarily "compete" the way it has in the past. > How much better will they have to be than standard Linux > clusters before they get any mainstream interest? What technical > features could they add that couldn't be added to a Linux > cluster? These are critical questions. The big question that needs to be answered (by Microsoft) is whether or not they need to displace or take over for Linux to be successful in this space. I believe the answer to this is "no". It might be worth engaging John West (InsideHPC.com), and a number of others in this conversation, as there are some good and informed views about this out there. > > Cordially, -- Joseph Landman, Ph.D Founder and CEO Scalable Informatics LLC, email: landman@scalableinformatics.com web : http://www.scalableinformatics.com http://jackrabbit.scalableinformatics.com phone: +1 734 786 8423 fax : +1 866 888 3112 cell : +1 734 612 4615 From jlforrest at berkeley.edu Wed Apr 2 16:01:58 2008 From: jlforrest at berkeley.edu (Jon Forrest) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <47F407A5.9090403@scalableinformatics.com> References: <47F402EC.3050701@berkeley.edu> <47F407A5.9090403@scalableinformatics.com> Message-ID: <47F41066.7030103@berkeley.edu> Joe Landman wrote: > Hi Jon > > Jon Forrest wrote: >> First of all, I like Microsoft, and I voluntarily use >> Vista as my desktop of choice. I've built and run the >> Windows environments for the top CS and Civil Engineering >> departments in the US, and I was the first to port >> Postgres to Windows NT. > > Postgres is nice. Did you do a native (ms C compiler) port or a cygwin > port? It's an interesting story, which is OT to Beowulf. I was working for Mike Stonebraker in his research group. Jolly Chen and Andrew Yu and just added SQL support to Postgres. Mike and I had been to Microsoft several times to see about them supporting us, and other research. I was very interested in Windows NT and very impressed by Microsoft. This was back in about 1994 or 1995. Anyway, I had NT 3.51 running on a 16MB 486 and I decided to try to port what was to become PostgreSQL. I used the MS Visual C compiler, not cygwin, which I'm not sure existed back then. The Postgres code was surprisingly easy to port, although the porting went much faster when Mike was able to get a 64MB MIPS machine donated. The big accomplishment was that I was able to run the infamous Wisconsin Benchmark. > Oddly enough, for a long time I have believed (and we have proposed > privately to Microsoft) ways in which I think it could make a > difference. In short, there are codes that are now and will always be > windows based, and that is fine. They may need to run on a cluster, and > they should be able to. What kind of reaction have you received from them? What code are you thinking of that is now and will always be Windows based? > It needs to adapt. But not necessarily "compete" the way it has in the > past. How can they adapt to an Open/Free Software world? > The big question that needs to be > answered (by Microsoft) is whether or not they need to displace or take > over for Linux to be successful in this space. I believe the answer to > this is "no". But how will they be able to make any money? > It might be worth engaging John West (InsideHPC.com), and a number of > others in this conversation, as there are some good and informed views > about this out there. I hope they read this thread. Cordially, -- Jon Forrest Research Computing Support College of Chemistry 173 Tan Hall University of California Berkeley Berkeley, CA 94720-1460 510-643-1032 jlforrest@berkeley.edu From dag at sonsorol.org Wed Apr 2 16:39:30 2008 From: dag at sonsorol.org (Chris Dagdigian) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <47F402EC.3050701@berkeley.edu> References: <47F402EC.3050701@berkeley.edu> Message-ID: <3219A57F-D431-4747-9614-9B5DEBBF6C96@sonsorol.org> I still think MS HPC has a shot in this space. Remember the size of our market is pretty small -- a few thousand installs would have a pretty significant impact in this space although within MS that may qualify as an embarrassing product failure. To me the market for MS HPC products is with commercial ISVs that sell software but are currently buried under infinite permutations of hardware and software they need to support. Setting up a cluster, managing the various communication stacks, application integration and batch scheduling is still quite a bit of work. With a MS shrink-wrap HPC product there is the possibility that an ISV can offload much of that responsibility to Microsoft - or more reasonably some other software/support/consulting shop that has the necessary domain experience to handle end-user application integration and usage questions. Right now for commercial ISVs the support question is non trivial and involves many different sets of compilers, MPI implementations, transport fabrics and cluster schedulers. Anyway, the possibility that commercial vendors of engineering/ scientific software could support MS HPC as a reasonable base that allows for a single point of contact (or finger pointing) for integration/usage issues is really the only thing I can come up with. In the market I'm familiar with (life science) there are all sorts of devices that could benefit from a shrink-wrapped compute solution. Even the smallest labs can get grants to purchase the latest $500K confocal microscope or next-generation DNA Sequencing box. Each of those devices can spew out a terabyte per day of raw data and many times that stuff needs to be post processed and distilled down into different forms. A nice little 8-core box running a shrink-wrap HPC product with a single support contact could find a nice little niche in non-datacenter areas where significant compute is needed nearby some other sort of dedicated instrument or device. Personally I'll stick with Unix and Grid Engine / LSF myself but I do see some areas, use-cases and markets where MS HPC could benefit. The window of opportunity is shrinking though. My $.02 of course. -Chris On Apr 2, 2008, at 6:04 PM, Jon Forrest wrote: > First of all, I like Microsoft, and I voluntarily use > Vista as my desktop of choice. I've built and run the > Windows environments for the top CS and Civil Engineering > departments in the US, and I was the first to port > Postgres to Windows NT. > > That said, I just don't see how Microsoft's HPC server > can succeed. I'm not saying this for technical reasons, > as I'm sure Microsoft, with enough work, can build > a clustering environment that will work just fine. > But, why would anybody buy a Windows cluster when > there are so many great clustering environments (e.g. > Rocks, Perceus, Unicluster Express, ...) and so many > cluster-related packages (*MPI, SGE, PBS, gcc, > Torque, ...) available for free? What's more, from > what I can see, there is very little non-Microsoft-sponsored > development going on in HPC computing. > > Microsoft recently announced (somewhere, I can't find it) > the availability of a test cluster for universities to > use for financial applications. I bet they get some > interest since many business schools use Windows, plus > the cluster is free. > > But, the question remains. How can Microsoft compete with free? > How much better will they have to be than standard Linux > clusters before they get any mainstream interest? What technical > features could they add that couldn't be added to a Linux > cluster? > > Cordially, > -- > Jon Forrest > Research Computing Support > College of Chemistry > 173 Tan Hall > University of California Berkeley > Berkeley, CA > 94720-1460 > 510-643-1032 > jlforrest@berkeley.edu > _______________________________________________ > Beowulf mailing list, Beowulf@beowulf.org > To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From Shainer at mellanox.com Wed Apr 2 16:55:53 2008 From: Shainer at mellanox.com (Gilad Shainer) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <3219A57F-D431-4747-9614-9B5DEBBF6C96@sonsorol.org> Message-ID: <9FA59C95FFCBB34EA5E42C1A8573784F0107F41D@mtiexch01.mti.com> > -----Original Message----- > From: beowulf-bounces@beowulf.org > [mailto:beowulf-bounces@beowulf.org] On Behalf Of Chris Dagdigian > Sent: Wednesday, April 02, 2008 4:40 PM > To: Beowulf Mailing List > Subject: Re: [Beowulf] How Can Microsoft's HPC Server Succeed? > > > I still think MS HPC has a shot in this space. > > Remember the size of our market is pretty small -- a few > thousand installs would have a pretty significant impact in > this space although within MS that may qualify as an > embarrassing product failure. > According to the latest reports from IDC and others, the HPC server market is 20% of the total server market, and it is fast growing segment. Will reach 30% in 2 years or less. The virtualization effect helps with this % increase as it reduces the number of servers in the EDC segment. You can also check IDC report on InfiniBand - it is growing fast, and one of the reasons is the growth of the HPC market. The HPC market is not small at all, and it is growing very fast. This is the reason driving the big vendors. Gilad. From landman at scalableinformatics.com Wed Apr 2 17:22:27 2008 From: landman at scalableinformatics.com (Joe Landman) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <9FA59C95FFCBB34EA5E42C1A8573784F0107F41D@mtiexch01.mti.com> References: <9FA59C95FFCBB34EA5E42C1A8573784F0107F41D@mtiexch01.mti.com> Message-ID: <47F42343.8080602@scalableinformatics.com> Gilad Shainer wrote: > > >> -----Original Message----- >> From: beowulf-bounces@beowulf.org >> [mailto:beowulf-bounces@beowulf.org] On Behalf Of Chris Dagdigian >> Sent: Wednesday, April 02, 2008 4:40 PM >> To: Beowulf Mailing List >> Subject: Re: [Beowulf] How Can Microsoft's HPC Server Succeed? >> >> >> I still think MS HPC has a shot in this space. >> >> Remember the size of our market is pretty small -- a few >> thousand installs would have a pretty significant impact in >> this space although within MS that may qualify as an >> embarrassing product failure. >> > > According to the latest reports from IDC and others, the HPC server > market is 20% of the total server market, and it is fast growing > segment. Will reach 30% in 2 years or less. The virtualization effect It is also worth noting that aside from HPC the server market was flat. HPC is a $15B USD market, which while the USD is dropping, I think the market growth is driven in part by the fact that HPC is fundamentally an enabling technology. > helps with this % increase as it reduces the number of servers in the > EDC segment. You can also check IDC report on InfiniBand - it is growing > fast, and one of the reasons is the growth of the HPC market. The HPC > market is not small at all, and it is growing very fast. This is the > reason driving the big vendors. I would expect storage and data center backbone to be driving IB and 10 GbE (in addition to clusters). -- Joseph Landman, Ph.D Founder and CEO Scalable Informatics LLC, email: landman@scalableinformatics.com web : http://www.scalableinformatics.com http://jackrabbit.scalableinformatics.com phone: +1 734 786 8423 fax : +1 866 888 3112 cell : +1 734 612 4615 From rgb at phy.duke.edu Wed Apr 2 17:41:14 2008 From: rgb at phy.duke.edu (Robert G. Brown) Date: Sat Jul 19 01:06:59 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <47F42343.8080602@scalableinformatics.com> References: <9FA59C95FFCBB34EA5E42C1A8573784F0107F41D@mtiexch01.mti.com> <47F42343.8080602@scalableinformatics.com> Message-ID: On Wed, 2 Apr 2008, Joe Landman wrote: >>> Remember the size of our market is pretty small -- a few thousand installs >>> would have a pretty significant impact in this space although within MS >>> that may qualify as an embarrassing product failure. >>> >> >> According to the latest reports from IDC and others, the HPC server >> market is 20% of the total server market, and it is fast growing >> segment. Will reach 30% in 2 years or less. The virtualization effect > > It is also worth noting that aside from HPC the server market was flat. HPC > is a $15B USD market, which while the USD is dropping, I think the market > growth is driven in part by the fact that HPC is fundamentally an enabling > technology. > >> helps with this % increase as it reduces the number of servers in the >> EDC segment. You can also check IDC report on InfiniBand - it is growing >> fast, and one of the reasons is the growth of the HPC market. The HPC >> market is not small at all, and it is growing very fast. This is the >> reason driving the big vendors. > > I would expect storage and data center backbone to be driving IB and 10 GbE > (in addition to clusters). It is a market that will be growing with the affluence of various parts of the world, as well. Silicon India published an article maybe yesterday projecting India as either the number two or number three global economy in a surprisingly short time -- behind China as number 1 and MAYBE behind the US as number 2. They do have the (human) numbers, and there is considerable entrepreneurial energy there these days, as the money they are bringing home from providing cheap remote programming and support is showing up invested in all different levels of industry. The interesting question will be whether or not Microsoft can win much of these emerging markets. These are not deep pockets, and linux is free (in addition to doing a better job of nearly everything, but free is pretty important in India in particular). rgb -- Robert G. Brown Phone(cell): 1-919-280-8443 Duke University Physics Dept, Box 90305 Durham, N.C. 27708-0305 Web: http://www.phy.duke.edu/~rgb Book of Lilith Website: http://www.phy.duke.edu/~rgb/Lilith/Lilith.php Lulu Bookstore: http://stores.lulu.com/store.php?fAcctID=877977 From kalpana0611 at gmail.com Wed Apr 2 19:24:39 2008 From: kalpana0611 at gmail.com (Cally K) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] Re: SMPs + One processor machines = Heterogeneous Cluster In-Reply-To: References: Message-ID: hi again I found this website from lam-mpi on heterogeneous cluster, and it kinda answers my question but just to be on the safe side, i would like to know the response from the mailing list users. http://www.lam-mpi.org/faq/category11.php3 On 4/3/08, Mark Kosmowski wrote: > > > Message: 8 > > Date: Wed, 2 Apr 2008 21:35:34 +0800 > > From: "Cally K" > > Subject: [Beowulf] SMPs + One processor machines = Heterogeneous > > Cluster > > To: beowulf@beowulf.org > > Message-ID: > > > > Content-Type: text/plain; charset="iso-8859-1" > > > > Hi guys, > > > > Thanks for the previous answers to my question, did not expect that > kinda > > flow of information and it proved to be helpful in my research. I have a > > different question and I hope I can get some nice feedback. > > > > I am dealing with heterogeneous cluster in my research. I am a novice > when > > it comes to networking, and have only been around with cluster for about > a > > year. Anyway, from my previous question, and after doing some reading, > can I > > say that dual cores and quad cores are known as SMPs.? > > > > And would 1 say that, a cluster that has machines with 1 processor ( > etc. > > Pentium 4 )and 2 SMPS( it can be quad or dual ) but all belonging to one > > family ( intel for example ) -- can I say that the cluster is a > > heterogeneous cluster. > > My limited (and perhaps incorrect) understanding is that SMP denotes > more than one processor, regardless the number of cores that the > processor has. > > Heterogenous, to me, implies a cluster of machines running different > operating systems and/or different architectures. > > As an example, my personal is three Opteron systems, each of which has > two single core CPU's, all running OpenSUSE 10.2 64-bit. This is a > homogenous cluster of SMP machines. I plan to add a dual core, single > CPU Athlon (64) system to the cluster at some point. This planned > machine would (in my opinion) not be SMP (although the more I think > about it, the more I think I am either wrong or in the minority camp > about SMP and cores) and the cluster would remain homogenous even if I > were to put OpenSUSE 11.0 (64 bit) on the system (the new OpenSUSE > release and my projected budget have similar timescales). > > Even if I were to use an Intel Pentium system of whatever nomenclature > that is instruction compatible with my existing Opterons (the > difficulty in immediately knowing whether a Pentium chip is 64-bit > capable or not is one of the reasons I am sticking with AMD) the > system would remain homogenous. > > If, however, I were to add a Sun SPARC machine as a node, now I have a > heterogenous cluster. Likewise, if my new Athlon (64) node used > Solaris (for Intel), the cluster would be heterogenous in my opinion. > > My rule of thumb for heterogeneity is whether the programs to be run > need to be recompiled for a given node. If all of the nodes can use > one single compile of all of the codes needed, the cluster is not > heterogenous. I guess this would mean that running a 32-bit program > (I want karma for minding my usage of code vs. program here! :) ) on a > mixture of 32-bit and 64-bit machines would be homogenous, but if on > the same cluster, the 32-bit machines ran a 32-bit program and the > 64-bit machines ran a 64-bit program then that cluster might be > considered heterogenous. > > I hope this at least sparks an interesting conversation, > > Mark Kosmowski > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.scyld.com/pipermail/beowulf/attachments/20080403/24512505/attachment.html From xclski at yahoo.com Wed Apr 2 20:34:42 2008 From: xclski at yahoo.com (Ellis Wilson) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] Re: SMPs + One processor machines = Heterogeneous Cluster Message-ID: <345779.94446.qm@web37902.mail.mud.yahoo.com> > need to be recompiled for a given node. If all of the nodes can use > one single compile of all of the codes needed, the cluster is not > heterogenous. I guess this would mean that running a 32-bit program > (I want karma for minding my usage of code vs. program here! :) ) on a > mixture of 32-bit and 64-bit machines would be homogenous, but if on > the same cluster, the 32-bit machines ran a 32-bit program and the > 64-bit machines ran a 64-bit program then that cluster might be I typically maintain a sub-environment, if one will, where I can run my 32 bit applications within a 64 bit operating system. All my nodes could do the same simply by chrooting in first prior to execution, and thus share exactly the same operating system capable of both 64 and 32 bit operation. Does one half of them running 32 bit applications and one half of them running 64 bit applications qualify them as heterogeneous even though their hardware and software are identical? I wouldn't say so. I typically associate heterogeneous with hardware, since, unlike software, it lacks the needed mutability to assimilate into a common architecture. Just my input, obviously I've not been at this long and would enjoy any suggestions otherwise. Ellis ____________________________________________________________________________________ You rock. That's why Blockbuster's offering you one month of Blockbuster Total Access, No Cost. http://tc.deals.yahoo.com/tc/blockbuster/text5.com From peter.st.john at gmail.com Wed Apr 2 21:40:40 2008 From: peter.st.john at gmail.com (Peter St. John) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] FREEBSD CLUSTER In-Reply-To: <73a01bf20803311114m502d9a85t4b35b09ecea78c1@mail.gmail.com> References: <5ce468b90803301012g1aeecb86h492c6cd266995a2e@mail.gmail.com> <73a01bf20803311114m502d9a85t4b35b09ecea78c1@mail.gmail.com> Message-ID: Slashdot mentioned a cluster topology based on "perfect difference sets" used to build EKA, http://en.wikipedia.org/wiki/EKA_%28supercomputer%29 but the wiki article doesn't mention it. The article http://punetech.com/building-eka-the-worlds-fastest-privately-funded-supercomputer/only mentions papers that require subscriptions. Peter On 3/31/08, Rayson Ho wrote: > > You can get detailed information on setting up a BSD HPC cluster at: > > http://people.freebsd.org/~brooks/papers/ > > Grid Engine was ported and maintained by the author... > > > Rayson > > > > > On Sun, Mar 30, 2008 at 12:12 PM, Israel Lehnen Silva > wrote: > > Hello friends! > > > > I am Brazilian, and i am doing a Beowulf Cluster in OS FreeBSD for a > > project of University where i study the networking Infrastructure of > > the link between the nodes. > > If someone has other material that can help me, send me for me to > > increase my project! > > > > Counting on the cooperation and support of the brothers. > > -- > > > > Att. Israel Lehnen Silva > > _______________________________________________ > > Beowulf mailing list, Beowulf@beowulf.org > > To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf > > > _______________________________________________ > Beowulf mailing list, Beowulf@beowulf.org > To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.scyld.com/pipermail/beowulf/attachments/20080403/691d359b/attachment.html From sigut at id.ethz.ch Wed Apr 2 22:18:11 2008 From: sigut at id.ethz.ch (G.M.Sigut) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] Re: "Code" vs. "Codes" In-Reply-To: <200804021416.m32EFYJv028051@bluewest.scyld.com> References: <200804021416.m32EFYJv028051@bluewest.scyld.com> Message-ID: <1207199891.6550.5.camel@gms2.ethz.ch> On Wed, 2008-04-02 at 07:16 -0700, beowulf-request@beowulf.org wrote: ... > Date: Tue, 1 Apr 2008 19:23:18 +0000 > From: "Andrew M.A. Cater" ... > Yes, but this is "code" as in "thing to keep other things secret" > > cryptography, cryptanalysis, "code breaking" > > "code" as in "I'm finding it hard to code this thing up / the code is > all in asm rather than C" is standard computer speak in the US and > English speaking world. > > "codes" as in "the weather forecasting / CFD / higher order topology > computer programs that I've been working on" - feels utterly alien > but then "code" and "codex" are cognate. Yes. There is also e.g. ASME Code as a set of rules for calculating stresses in mechanical parts... So when I say "I code code code" I am not having a strange hiccup, but I mean "I am writing (coding) a program (code) implementing stress calculation prescription (code)". I always thought that the usage was slightly strange. George -- >>>>>>>>>>>>>>>>>>>>>>>>>> George M. Sigut <<<<<<<<<<<<<<<<<<<<<<<<<< ETH Zurich, Informatikdienste, Abteilung Systemdienste, CH-8092 Zurich Swiss Federal Inst. of Technology Zurich, IT Services, System Services e-mail: sigut@id.ethz.ch, Phone:+41 44 632 5763, Fax: +41 44 632 1022 >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>-<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< From rgb at phy.duke.edu Thu Apr 3 04:03:54 2008 From: rgb at phy.duke.edu (Robert G. Brown) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] Re: "Code" vs. "Codes" In-Reply-To: <1207199891.6550.5.camel@gms2.ethz.ch> References: <200804021416.m32EFYJv028051@bluewest.scyld.com> <1207199891.6550.5.camel@gms2.ethz.ch> Message-ID: On Thu, 3 Apr 2008, G.M.Sigut wrote: > So when I say "I code code code" I am not having a strange hiccup, > but I mean "I am writing (coding) a program (code) implementing > stress calculation prescription (code)". I always thought that the > usage was slightly strange. Don't forget to encrypt the final result according to the law. "I code code code I've just coded according to the code." But this is hardly the only English word with a half dozen meanings that span verb and noun...;-) rgb -- Robert G. Brown Phone(cell): 1-919-280-8443 Duke University Physics Dept, Box 90305 Durham, N.C. 27708-0305 Web: http://www.phy.duke.edu/~rgb Book of Lilith Website: http://www.phy.duke.edu/~rgb/Lilith/Lilith.php Lulu Bookstore: http://stores.lulu.com/store.php?fAcctID=877977 From deadline at eadline.org Thu Apr 3 06:01:19 2008 From: deadline at eadline.org (Douglas Eadline) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] FREEBSD CLUSTER In-Reply-To: References: <5ce468b90803301012g1aeecb86h492c6cd266995a2e@mail.gmail.com> <73a01bf20803311114m502d9a85t4b35b09ecea78c1@mail.gmail.com> Message-ID: <58166.192.168.1.213.1207227679.squirrel@mail.eadline.org> >From the EKA Wikipedia article: "The CRL supercomputer has been built using CLOS architecture" Maybe a link to: http://en.wikipedia.org/wiki/Clos_network would help the article. I am not sure what they mean by "perfect difference sets". Anyone know? -- Doug > Slashdot mentioned a cluster topology based on "perfect difference sets" > used to build EKA, http://en.wikipedia.org/wiki/EKA_%28supercomputer%29 > but > the wiki article doesn't mention it. The article > http://punetech.com/building-eka-the-worlds-fastest-privately-funded-supercomputer/only > mentions papers that require subscriptions. > Peter > > On 3/31/08, Rayson Ho wrote: >> >> You can get detailed information on setting up a BSD HPC cluster at: >> >> http://people.freebsd.org/~brooks/papers/ >> >> Grid Engine was ported and maintained by the author... >> >> >> Rayson >> >> >> >> >> On Sun, Mar 30, 2008 at 12:12 PM, Israel Lehnen Silva >> wrote: >> > Hello friends! >> > >> > I am Brazilian, and i am doing a Beowulf Cluster in OS FreeBSD for a >> > project of University where i study the networking Infrastructure of >> > the link between the nodes. >> > If someone has other material that can help me, send me for me to >> > increase my project! >> > >> > Counting on the cooperation and support of the brothers. >> > -- >> > >> > Att. Israel Lehnen Silva >> > _______________________________________________ >> > Beowulf mailing list, Beowulf@beowulf.org >> > To change your subscription (digest mode or unsubscribe) visit >> http://www.beowulf.org/mailman/listinfo/beowulf >> > >> _______________________________________________ >> Beowulf mailing list, Beowulf@beowulf.org >> To change your subscription (digest mode or unsubscribe) visit >> http://www.beowulf.org/mailman/listinfo/beowulf >> > > > !DSPAM:47f4612e224331745845678! > _______________________________________________ > Beowulf mailing list, Beowulf@beowulf.org > To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf > > > !DSPAM:47f4612e224331745845678! > -- Doug From gbyshenk at byshenk.net Wed Apr 2 15:44:03 2008 From: gbyshenk at byshenk.net (Greg Byshenk) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <47F402EC.3050701@berkeley.edu> References: <47F402EC.3050701@berkeley.edu> Message-ID: <20080402224403.GY90095@gby1.aoes.com> On Wed, Apr 02, 2008 at 03:04:28PM -0700, Jon Forrest wrote: > But, the question remains. How can Microsoft compete with free? > How much better will they have to be than standard Linux > clusters before they get any mainstream interest? What technical > features could they add that couldn't be added to a Linux > cluster? The thing to remember is that a cluster (even one running "free" -- as in beer -- software) is not without cost. Apart from hardware, licenses (if required), etc., a business will also have to find (and pay) someone to build and maintain the cluster. A former co-worker of mine now works elsewhere, at a company where they are looking to build a cluster for some research he will be doing, and it will almost certainly run Windows. And this is not because there is something "better" about Windows, but because they are a Windows shop, and thus the cost of hiring or training staff would more than outweigh any savings on licenses. -- Greg Byshenk - gbyshenk@byshenk.net - Leiden, NL From joshua_mora at usa.net Wed Apr 2 21:07:30 2008 From: joshua_mora at usa.net (Joshua mora acosta) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] HPL Benchmarking and Optimization Message-ID: <512mDcegE0136S07.1207195650@cmsweb07.cms.usa.net> Get for AMD based systems ACML and gcc,pgi or pathscale Get for Intel based systems MKL and intel compiler run N problem size around 90% workload.is, 1.8GB per core memory footprint. Run NB 192 on AMD, I don't know the best blocking factor for MKL. I've tried the same 192 and does fairly well. Set affinity for the mpi even with 1 socket runs. Run PxQ 2x2,2x4,4x4,.. depending on the number of cores. With the above you should get on AMD and on Intel at least 77% efficiency. As suggested by Tom, Goto library will give you good performance as well. You can try also the multithreaded version so use PxQ=1x1 and OMP_NUM_THREADS=4 for a single socket quadcore. Reduce misses with huge pages. If you get below 75% efficiency, you are doing something wrong. If you do more than 85% on quadcore, please let me know :) Regards, Joshua ------ Original Message ------ Received: Wed, 02 Apr 2008 12:33:25 PM PDT From: Ellis Wilson To: beowulf@beowulf.org Subject: Re: [Beowulf] HPL Benchmarking and Optimization > Ellis Wilson wrote: > > Currently I get these kind of numbers from tested > > computers using the > > same environment (gentoo, fortran in gcc, hpl, all > > same compilation > > options): > > 1 x Core2Duo (2.1ghz/core, 2gigs ram) - 2.3Gflops > > 1 x Athlon 64 3500+ (2.2ghz, 1gig ram) - 1.0Glops > > 4 x Core2Duo (2.1ghz/core for a total of 8 cores, > > 2gigs ram/node, > > 100mbit Ethernet interconnect) - 6.7Gflops > > Sorry to double post all, however, I realized my issue > involved running > HPL on the reference library of BLAS that is generic > for every > architecture and didn't want to waste anyones time. > Giving Portage the > benefit of the doubt, I had failed to check that it's > dependencies were > best for HPL. Following an install of ATLAS and > relinking to its > libraries, I've gotten the following numbers: > 1 x Athlon64 3500+ (2.2ghz, 1gig ram) - 3.6GFlops > 1 x Phenom9600 Quadcore (2.3ghz/core, 2gigs ram) - > 11.9GFlops > > I'll likely try MKL soon for the Intel processors I'm > interested in. > > The phenom9600 had previously only gotten 4.5 GFlops, > and when I tested > it the second time I simply used the same environment > I had compiled for > the athlon64. Certainly compiling ATLAS native on the > phenom will > increase the result, hopefully about 350% like with > the athlon64 (though > I suspect things will be interesting due to bandwidth, > etc for quadcores). > > Anyway, not to end the thread I still am wondering: > > Do those of you who have professional installations or > even simply large > setups that are unsure of the exact code which will be > run upon your > cluster utilize compilation options such as -O3, > funroll-loops, > -fomit-frame-pointer, etc? > > Thanks, > > Ellis > > > > > ____________________________________________________________________________________ > You rock. That's why Blockbuster's offering you one month of Blockbuster Total Access, No Cost. > http://tc.deals.yahoo.com/tc/blockbuster/text5.com > _______________________________________________ > Beowulf mailing list, Beowulf@beowulf.org > To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf > From rreis at aero.ist.utl.pt Thu Apr 3 01:16:25 2008 From: rreis at aero.ist.utl.pt (Ricardo Reis) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] Re: "Code" vs. "Codes" In-Reply-To: <1207199891.6550.5.camel@gms2.ethz.ch> References: <200804021416.m32EFYJv028051@bluewest.scyld.com> <1207199891.6550.5.camel@gms2.ethz.ch> Message-ID: What is a program? Isn't it a codified algorithm? So whats the worry? By the way, when I started more into CFD I felt strange by people refering to programs as "c?digos" (codes) instead of programs and always thought it was an easy influence from the english language (I'm in Portugal) so go figure :) Ricardo Reis 'Non Serviam' PhD student @ Lasef Computational Fluid Dynamics, High Performance Computing, Turbulence http://www.lasef.ist.utl.pt & Cultural Instigator @ R?dio Zero http://www.radiozero.pt http://www.flickr.com/photos/rreis/ From kekechen at cc.gatech.edu Wed Apr 2 21:18:30 2008 From: kekechen at cc.gatech.edu (Keke Chen) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] 3rd CFP -- CoopIS08 Message-ID: <47F45A96.2090709@cc.gatech.edu> We apologize if you receive multiple copies ======== 3nd Call For Papers =================== 16th International Conference on COOPERATIVE INFORMATION SYSTEMS (CoopIS 2008) Monterrey, Mexico, Nov 12 - 14, 2008 http://www.cs.rmit.edu.au/fedconf Acceptance rate of CoopIS in recent years was approx. 20% Cooperative Information Systems are the cornerstone for moving the technical network infrastructure to a meaningful integrated information infrastructure. The CIS paradigm has traditionally encompassed distributed systems technologies such as middleware, business process management (BPM) and Web technologies. In recent years service oriented architectures have fundamentally altered the technological landscape of CIS systems. Service Oriented Computing (SOC) introduces the service abstraction (a remotely accessible software component) as the building block of both inter and intra organizational distributed applications and its supporting middleware. Cooperative Information Systems applications are heavily distributed and highly coordinated, often exhibiting inter-organizational interaction patterns and requiring distributed access and sharing of computing and information resources. Typically they fall under the categories of e-Business, e-Commerce, e-Government, e-Health, e-Science among others. The CoopIS conference series has established itself as a major international forum for exchanging ideas and results on scientific research for practitioners in fields such as computer supported cooperative work (CSCW), middleware, Internet data management, electronic commerce, human-computer interaction, workflow management, agent technologies, and software architectures, to name a few. In addition, the 2008 edition of CoopIs aims to highlight the impact of service oriented computing and the importance of sustainability of CIS as a necessary prerequisite for mission critical applications. As in previous years, CoopIS'08 will be part of a joint event with other conferences, in the context of the OTM ("On The Move") federated conferences, covering different aspects of distributed information systems. Topics that are addressed by CoopIS'08 are logically grouped in three broad areas, and include but are not limited to: * Business Process Management and Compliance o Business Process Integration and Management o Cooperation Aspects in Business Process Management o Distributed Workflow Management and Systems o Service orchestration and service compositions o Process choreographies o Business process compliance o Integrated supply chains o Concurrent engineering and distributed groupware o Business level policies o Governance, risk and compliance models and runtimes o Sustainability of processes * Advanced middleware and architectures and runtimes o Service oriented middleware o Web services standards and runtimes o Grid computing infrastructure o Enterprise Grids architectures and services o Web centric information and processing architectures o Semantic interoperability o Self-adapting and self-healing systems o Model driven middleware architectures o Multi-agent systems and architectures for CIS o Peer-to-peer technologies o Security and privacy in CIS o Quality of service in cooperative information systems o Mediation, matchmaking, and brokering architectures o Collaboration and negotiation protocols o Markets, auctions, exchanges, and coalitions * CIS Applications o Novel CIS applications for the large organizations: e-business, e-commerce, e-government o Advances in e-science and Grid computing applications o Medical and biological information systems o Industrial applications of CIS o Web 2.0 IMPORTANT DATES Abstract Submission Deadline June 8, 2008 Paper Submission Deadline June 15, 2008 Acceptance Notification August 10, 2008 Camera Ready Due August 25, 2008 Registration Due August 25, 2008 OTM Conferences November 9 - 14, 2008 SUBMISSION GUIDELINES Papers submitted to CoopIS'08 must not have been accepted for publication elsewhere or be under review for another workshop or conference. All submitted papers will be carefully evaluated based on originality, significance, technical soundness, and clarity of expression. All papers will be refereed by at least three members of the program committee, and at least two will be experts from industry in the case of practice reports. All submissions must be in English. Submissions must not exceed 18 pages in the final camera-ready paper style. Submissions must be laid out according to the final camera-ready formatting instructions and must be submitted in PDF format. The paper submission site will be announced later Failure to comply with the formatting instructions for submitted papers will lead to the outright rejection of the paper without review. Failure to commit to presentation at the conference automatically excludes a paper from the proceedings. ORGANISATION COMMITTEE General Co-Chairs * Robert Meersman, VU Brussels, Belgium * Zahir Tari, RMIT University, Australia Program Committee Co-Chairs * Johann Eder, University of Klagenfurt, Austria * Masaru Kitsuregawa, University of Tokyo, Japan * Ling Liu, Georgia Institute of Technology, USA Program Committee Members (to be extended and confirmed) * Ghaleb Abdulla, Lawrence Livermore National Laboratory, USA * Marco Aiello, University of Groningen, The Netherlands * Joonsoo Bae, Chonbuk National Universiry, South Korea * Alistair Barros, SAP, Research Centre Brisbane, Australia * Zohra Bellahsene, LIRMM- CNRS/Universit? Montpellier 2, France * Salima Benbernou, University Lyon 1, France * Djamal Benslimane, University of Lyon, France * M. Brian Blake, Georgetown University, Washington DC, USA * Klemens B?hm, University of Karlsruhe, Germany * Christoph Bussler, Cisco Systems, Inc, USA * Ying Cai, Iowa State University, USA * James Caverlee, Texas A&M University, USA * Keke Chen, Yahoo!, USA * Vincenzo D'Andrea, University of Trento, Italy * Umesh Dayal, HP Labs * Xiaoyong Du, Renmin University of China, PR China * Marlon Dumas, University of Tartu, Estonia * Schahram Dustdar, Vienna University of Technology, Austria * Rik Eshuis, Eindhoven University, The Netherlands * Opher Etzion, IBM Israel Software Lab * Renato Fileto, Federal University of Santa Catarina, Brazil * Klaus Fischer, DFKI, Germany * Avigdor Gal, Technion Israel Institute of Technology, Israel * Bugra Gedik, IBM TJ Watson, USA * Dimitrios Georgakopoulos, Telcordia, USA * Paul Grefen, Eindhoven University of Technology, The Netherlands * Amarnath Gupta, University of California San Diego, USA * Mohand-Said Hacid, Lyon University, France * Thorsten Hampel, University of Paderborn, Germany * Geert-Jan Houben, TU Eindhoven & VUB Brussels * Richard Hull, Lucent Technologies, USA * Patrick Hung, University of Ontario Institute of Technology (UOIT), Canada * Paul Johannesson, Royal Institute of Technology (KTH), Sweden * Dimka Karastoyanova, University of Stuttgart, Germany * Rania Khalaf, IBM Research * Hiroyuki Kitagawa, University of Tsukuba * Shim Kyusock, Seoul National Univ. * Akhil Kumar, Penn State University, USA * Wang-Chien Lee, Pennsylvania State University, USA * Frank Leymann, University of Stuttgart, Germany * Chen Li, University of California, Irvine, USA * Sanjay K. Madria, Missouri University of Science and Technology, USA * Leo Mark, Georgia Institute of Technology * Maristella Matera, DEI - Politecnico di Milano, Italy * Massimo Mecella, Universita' di Roma, Italy * Nirmal Mukhi, IBM T J Watson Research Center * Mohamed Mokbel, University of Minnessota, USA * J?rg M?ller, Technische Universit?t Clausthal * Miyuki Nakano, University of Tokyo, Japan * Moira Norrie, ETH Zurich, Switzerland * Werner Nutt, Free University of Bozen-Bolzano, Italy * Andreas Oberweis, University of Karlsruhe, Germany * Cesare Pautasso, University of Lugano, Switzerland * Barbara Pernici, Politecnico di Milano, Italy * Frank Puhlmann, Hasso Plattner Institut, Germany * Manfred Reichert, Ulm University, Germany * Stefanie Rinderle-Ma, Ulm University, Germany * Lakshmish Ramaswamy, University of Georgia, USA * Duncan Ruiz, Catholic University of RS, Brazil * Kai-Uwe Sattler, TU Ilmenau, Germany * Ralf Schenkel, Max-Planck-Institut Informatik, Germany * Jialie Shen, Singapore Management University, Singapore * Aameek Singh, IBM Almaden Research Center * Mudhakar Srivatsa, IBM TJ Watson Research Center, USA * Jianwen Su, University of California, Santa Barbara, USA * Wei Tang, Teradata Corp. USA * Anthony Tung, National University of Singapore, Singapore * Susan Urban, Texas Tech University, USA * Willem-Jan Van den Heuvel, Tilburg University, The Netherlands * Maria Esther Vidal, Universidad Simon Bolivar, Caracas Venezuela * Shan Wang, Renmin University of China, PR China * X. Sean Wang, University of Vermont, USA * Jeffrey Yu, Chinese University of Hong Kong, HK * Matthias Weske, University of Potsdam, Germany * Li Xiong, Emory University, USA * Jian Yang, Macquarie University, Australia * Masatoshi Yoshikawa, Kyoto University, Japan * Leon Zhao, University of Arizona, USA * Xiaofang Zhou, University of Queensland, Australia * Aoying Zhou, East China Normal University, PR China * Michael zur Muehlen, Stevens Institute of Technology, USA From kekechen at cc.gatech.edu Wed Apr 2 21:20:43 2008 From: kekechen at cc.gatech.edu (Keke Chen) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] 3rd CFP -- GADA 08: Intl Conf. on Grid computing, high-performAnce and Distributed Applications Message-ID: <47F45B1B.8060602@cc.gatech.edu> We apologize if you receive multiple copies ======== 3nd Call For Papers =================== International Conference on Grid computing, high-performAnce and Distributed Applications (GADA'08) Monterrey, Mexico, Nov 13 - 14, 2008 http://www.cs.rmit.edu.au/fedconf In the last decade, grid computing has developed into one of the most important topics in the computing field. The research area of grid computing has been making particularly rapid progress in the last few years, due to the increasing number of scientific applications that are demanding intensive use of computational resources and a dynamic and heterogeneous infrastructure. Within this framework, the GADA workshop arose in 2004 as a forum for researchers in grid computing whose aim was to extend their background in this area, and more specifically, for those who used grid environments in managing and analyzing data. Both GADA'04 and GADA'05 were constituted as successful events, due to the large number of high-quality papers received, as well as the brainstorming of experiences and ideas interchanged in the associated forums. Because of this demonstrated success, GADA was upgraded as a Conference within On The Move Federated Conferences and Workshops (OTM'06). GADA'06 covered a broader set of disciplines, although grid computing kept a key role in the set of main topics of the conference. The objective of grid computing is the integration of heterogeneous computing systems and data resources with the aim of providing a global computing space. The achievement of this goal is creating revolutionary changes in the field of computation, because it enables resource sharing across networks, with data being one of the most important resources. Thus, data access, management and analysis within grid and distributed environments are also dealt as main part of the conference. Therefore, the main goal of GADA'08 is to provide a framework in which a community of researchers, developers and users can exchange ideas and works related to grid, high-performance and distributed applications and systems. The second goal of GADA'08 is to create interaction between grid computing researchers and the other OTM attendees. GADA'08 intends to draw a highly diverse body of researchers and practitioners by being part of the "On the Move to Meaningful Internet Systems and Ubiquitous Computing 2008" federated conferences event that includes five co-located conferences: * GADA'08 (International Conference on Grid computing, high-performAnce and Distributed Applications) * CoopIS'08 (International Conference on Cooperative Information Systems) * DOA'08 (International Symposium on Distributed Objects and Applications) * ODBASE'08 (International Conference on Ontologies, DataBases, and applications of Semantics) * IS'08 (Information Security Symposium) TOPICS OF INTEREST Topics of interest include, but are not limited to: * Computational grids * Data grids * High-performance computing * Distributed applications * Cluster computing * Parallel applications * Grid infrastructures for data analysis * High-performance computing for data-intensive applications * Grid computing infrastructures, middleware and tools * Mobile Grid Computing * Grid computing services * Collaboration technologies * Data analysis and management on grids * Distributed and parallel I/O systems * Extracting knowledge from data grids * Agent architectures for grid and distributed environments * Agent-based data extraction in distributed systems * Semantic Grid * Security in distributed environments * Security in computational and data grids * Grid standards as related to applications IMPORTANT DATES Abstract Submission Deadline June 8, 2008 Paper Submission Deadline June 15, 2008 Acceptance Notification August 10, 2008 Camera Ready Due August 25, 2008 Registration Due August 25, 2008 OTM Conferences November 9 - 14, 2008 SUBMISSION GUIDELINES Papers submitted to GADA'08 must not have been accepted for publication elsewhere or be under review for another workshop or conference. All submitted papers will be carefully evaluated based on originality, significance, technical soundness, and clarity of expression. All submissions must be in English. Submissions should be in PDF format and must not exceed 18 pages in the final camera-ready format. The paper submission site will be announced shortly Failure to commit to presentation at the conference automatically excludes a paper from the proceedings. GADA PC co-chairs * Dennis Gannon Computer Science Department Indiana University Lindley Hall, Room 215 150 S. Woodlawn Ave. Bloomington, IN 47405-7104 Phone: (812) 855-5184 Fax: (812) 855-4829 Email: gannon@cs.indiana.edu * Pilar Herrero Facultad de Inform?tica Universidad Polit?cnica de Madrid Madrid (Spain) Phone: (+34) 91.336.74.56 Fax: (+34) 91.336.65.95E Email: pherrero@fi.upm.es * Daniel S. Katz Louisiana State University Louisiana (USA) Phone: (+1) 225.578.2750 Fax: (+1) 225.578.5362 Email: d.katz@ieee.org * Mar?a S. P?rez Facultad de Inform?tica Universidad Polit?cnica de Madrid Madrid (Spain) Phone: (+34) 91.336.73.80 Fax: (+34) 91.336.73.73 Email: mperez@fi.upm.es Program Committee (to be confirmed and extended) * Adam Wierzbicki, Polish-Japanese Institute of Information Technology, Poland * Akshai Aggarwal, University of Windsor, Canada * Alan Sussman, University of Maryland, College Park, USA * Alberto Sanchez, UPM, Spain * Anastasios Gounaris, Aristotle University of Thessaloniki, Greece * Artur Andrzejak, Zuse Institute Berlin (ZIB), Germany * Beniamino Di Martino, Department of Information Engineering, Seconda Universit? di Napoli, Italy * Bhanu Prasad, Florida A &M University, USA * Blanca Caminero Herraez, Universidad de Castilla-La Mancha, Spain * Carmela Comito, University of Calabria, Italy * Cho-Li Wang, Hong Kong University, China * Costin Badica, University of Craiova, Romania * Dana Petcu, Western University of Timisoara, Romania * Edgar Magana, CISCO Systems, USA * Eduardo Huedo, Universidad Complutense de Madrid, Spain * Elghazali Talbi, University of Lille, France * Ewa Deelman, USC Information Sciences Institute, USA * Felix Garc?a, Universidad Carlos III, Spain * F?lix J. Garc?a Clemente , Universidad de Murcia, Spain * Francisco Jos? da Silva e Silva, Universidade Federal do Maranh?o, Brasil * Francisco Luna, University of Malaga, Spain * Geoff Coulson,, Lancaster University, UK * Gregorio Martinez, Universidad de Murcia, Spain * Hamid Sarbazi-Azad, Sharif University of Technology, Iran * Heinz Stockinger, Swiss Institute of Bioinformatics, Lausanne, Switzerland * Hong Ong, Oak Ridge National Laboratory, USA * Ignacio M. Llorente, UCM-CAB, Madrid, Spain * Jemal Abawajy, Deakin University, Victoria, Australia * Jes?s Carretero, Universidad Carlos III, Spain * Jinjun Chen , Swinburne University of Technology, Australia * Jordi Torres, Barcelona SuperComputing Center (BSC-CNS), Spain * Jose Cunha, Universidade Nova de Lisboa, Portugal * Jose L. Bosque, Universidad de Cantabria, Spain * Jos? Luis V?zquez Poletti , Universidad Complutense de Madrid, Spain * Jose M. Pe?a, UPM, Spain * Juan A. Bot?a Blaya, Universidad de Murcia, Spain * Kamil Kuliberda, Polish-Japanese Institute of Information Technology, Poland * Kostas Karasavvas, National e-Science Centre, UK * Laurent Lefevre, INRIA, France * Manish Parashar, Rutgers University, NJ * Manuel Salvadores, University of Southampton, UK * Marcin Paprzycki, Systems Research Institute Polish Academy of Science, Poland * Maria Ganzha, Elblag University of Humanities and Economy, Poland * Mario Cannataro, Univ. of Catanzaro, Italy * Marios Dikaiakos, University of Cyprus, Cyprus * Mark Baker , University of Reading, UK * Markus Endler, PUC-Rio, * Mirela Notare, Barddal University, Brazil * Neil P Chue Hong , The University of Edinburgh, UK * Oscar Ardaiz, Universidad de Navarra, Spain * Pascal Bouvry, Universit? du Luxembourg, Luxembourg * Rajkumar Buyya, University of Melbourne, Melbourne, Australia * Reagan Moore, San Diego Supercomputer Center (SDSC), USA * Rizos Sakellariou, Univ. of Manchester, UK * Rosa M. Badia, UPC, Barcelona, Spain * Ruben S. Montero, UCM-CAB, Madrid, Spain * Santi Caball? Llobet, Open University of Catalonia, Spain * Sattar B. Sadkhan Almaliky, Iraq - Alnahrain University, Iraq * Toni Cortes, UPC, Barcelona, Spain * V?ctor Robles, UPM, Spain * Geoffrey Fox, Indiana University, USA * Shantenu Jha, Louisiana State University, USA * Alfredo Cuzzocrea, University of Calabria, Italy * Liviu Joita, University of Oxford, UK From 3lucid at gmail.com Thu Apr 3 12:52:54 2008 From: 3lucid at gmail.com (Kyle Spaans) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <20080402224403.GY90095@gby1.aoes.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> Message-ID: <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> On Wed, Apr 2, 2008 at 7:39 PM, Chris Dagdigian wrote: > spew out a terabyte per day of raw data and many times that stuff needs to > be post processed and distilled down into different forms. A nice little > 8-core box running a shrink-wrap HPC product with a single support contact > could find a nice little niche in non-datacenter areas where significant > compute is needed nearby some other sort of dedicated instrument or device. ... On Wed, Apr 2, 2008 at 6:44 PM, Greg Byshenk wrote: > a business will also have to find > (and pay) someone to build and maintain the cluster. Forgive me perhaps for being naive, but why can't a knowledgeable teenager / college student be paid ~$10/hour plus on-call time to do a setup like this? Presuming they only need to hire someone to do setup/administration/support (and not the actual programming itself). From peter.st.john at gmail.com Thu Apr 3 13:02:17 2008 From: peter.st.john at gmail.com (Peter St. John) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] FREEBSD CLUSTER In-Reply-To: <58166.192.168.1.213.1207227679.squirrel@mail.eadline.org> References: <5ce468b90803301012g1aeecb86h492c6cd266995a2e@mail.gmail.com> <73a01bf20803311114m502d9a85t4b35b09ecea78c1@mail.gmail.com> <58166.192.168.1.213.1207227679.squirrel@mail.eadline.org> Message-ID: Perfect Difference Sets (I just learned) are a math thing; a set of K+1 numbers (K turns out to be an interesting parameter) from the N-1 nonzero residues 1, 2, 3, ..., N-1 of integers modulo N, so that that every residue is the difference of some pair of numbers in the PD set. If N = K^2 + K + 1 and K is a power of two it works out, e.g. K = 2, N = 7 (a prime, so the residues form a finite field, also) e.g. {1, 2, 4} is a PDS mod 7, e.g. 2-4 = -2 = 5 (mod 7). So all of the 6 nonzero residues can be produced by differences of those 3 in the PDS. But how this is used as a coding for a network topology I have no idea. The Wolfram site deines PDS as above but doesn't mention networks. Peter On 4/3/08, Douglas Eadline wrote: > > > From the EKA Wikipedia article: > > "The CRL supercomputer has been built using CLOS architecture" > > Maybe a link to: > > http://en.wikipedia.org/wiki/Clos_network > > would help the article. I am not sure what they mean > by "perfect difference sets". Anyone know? > > -- > Doug > > > > > > Slashdot mentioned a cluster topology based on "perfect difference sets" > > used to build EKA, http://en.wikipedia.org/wiki/EKA_%28supercomputer%29 > > but > > the wiki article doesn't mention it. The article > > > http://punetech.com/building-eka-the-worlds-fastest-privately-funded-supercomputer/only > > mentions papers that require subscriptions. > > Peter > > > > On 3/31/08, Rayson Ho wrote: > >> > >> You can get detailed information on setting up a BSD HPC cluster at: > >> > >> http://people.freebsd.org/~brooks/papers/ > >> > >> Grid Engine was ported and maintained by the author... > >> > >> > >> Rayson > >> > >> > >> > >> > >> On Sun, Mar 30, 2008 at 12:12 PM, Israel Lehnen Silva > >> wrote: > >> > Hello friends! > >> > > >> > I am Brazilian, and i am doing a Beowulf Cluster in OS FreeBSD for > a > >> > project of University where i study the networking Infrastructure of > >> > the link between the nodes. > >> > If someone has other material that can help me, send me for me to > >> > increase my project! > >> > > >> > Counting on the cooperation and support of the brothers. > >> > -- > >> > > >> > Att. Israel Lehnen Silva > >> > _______________________________________________ > >> > Beowulf mailing list, Beowulf@beowulf.org > >> > To change your subscription (digest mode or unsubscribe) visit > >> http://www.beowulf.org/mailman/listinfo/beowulf > >> > > >> _______________________________________________ > >> Beowulf mailing list, Beowulf@beowulf.org > >> To change your subscription (digest mode or unsubscribe) visit > >> http://www.beowulf.org/mailman/listinfo/beowulf > >> > > > > > > > !DSPAM:47f4612e224331745845678! > > > _______________________________________________ > > Beowulf mailing list, Beowulf@beowulf.org > > To change your subscription (digest mode or unsubscribe) visit > > http://www.beowulf.org/mailman/listinfo/beowulf > > > > > > > !DSPAM:47f4612e224331745845678! > > > > > > -- > Doug > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.scyld.com/pipermail/beowulf/attachments/20080403/a07b7838/attachment.html From atchley at myri.com Thu Apr 3 13:13:13 2008 From: atchley at myri.com (Scott Atchley) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> Message-ID: <06F64C9C-BA0E-4B77-910A-AB907D886A3E@myri.com> On Apr 3, 2008, at 3:52 PM, Kyle Spaans wrote: > On Wed, Apr 2, 2008 at 7:39 PM, Chris Dagdigian > wrote: >> spew out a terabyte per day of raw data and many times that stuff >> needs to >> be post processed and distilled down into different forms. A nice >> little >> 8-core box running a shrink-wrap HPC product with a single support >> contact >> could find a nice little niche in non-datacenter areas where >> significant >> compute is needed nearby some other sort of dedicated instrument >> or device. > ... > On Wed, Apr 2, 2008 at 6:44 PM, Greg Byshenk > wrote: >> a business will also have to find >> (and pay) someone to build and maintain the cluster. > > > Forgive me perhaps for being naive, but why can't a knowledgeable > teenager / college student be paid ~$10/hour plus on-call time to do a > setup like this? Presuming they only need to hire someone to do > setup/administration/support (and not the actual programming itself). What if your data is sensitive or has HIPPA requirements? Do you want a part-timer having admin control of that data (regardless if it is Linux, Windows, or MacOSX)? Scott From 3lucid at gmail.com Thu Apr 3 13:16:26 2008 From: 3lucid at gmail.com (Kyle Spaans) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> Message-ID: <5a1205b30804031316r1e8f190bs2fd4ad79f7812194@mail.gmail.com> On Thu, Apr 3, 2008 at 3:52 PM, Kyle Spaans <3lucid@gmail.com> wrote: > Forgive me perhaps for being naive, but why can't a knowledgeable > teenager / college student be paid ~$10/hour plus on-call time to do a > setup like this? Presuming they only need to hire someone to do > setup/administration/support (and not the actual programming itself). Sorry, I should elaborate on that to keep this on-topic. FLOSS or Windows cluster, I think the idea is the same. But then again, how many people in that age demographic would have ever had access to Windows Clustering software? Wouldn't that make it more likely that you'd pay more for an admin for the Windows Cluster? That, appart from access to source code and that kind of thing, is the only problem I can think of. From 3lucid at gmail.com Thu Apr 3 13:18:37 2008 From: 3lucid at gmail.com (Kyle Spaans) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <06F64C9C-BA0E-4B77-910A-AB907D886A3E@myri.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> <06F64C9C-BA0E-4B77-910A-AB907D886A3E@myri.com> Message-ID: <5a1205b30804031318p656b9d06hc55d891f95ab1b4d@mail.gmail.com> On Thu, Apr 3, 2008 at 4:13 PM, Scott Atchley wrote: > On Apr 3, 2008, at 3:52 PM, Kyle Spaans wrote: > > Forgive me perhaps for being naive, but why can't a knowledgeable > > teenager / college student be paid ~$10/hour plus on-call time to do a > > setup like this? Presuming they only need to hire someone to do > > setup/administration/support (and not the actual programming itself). > > > > What if your data is sensitive or has HIPPA requirements? Do you want a > part-timer having admin control of that data (regardless if it is Linux, > Windows, or MacOSX)? Isn't that what NDAs are for? From gerry.creager at tamu.edu Thu Apr 3 13:30:43 2008 From: gerry.creager at tamu.edu (Gerry Creager) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> Message-ID: <47F53E73.5010507@tamu.edu> Kyle Spaans wrote: > On Wed, Apr 2, 2008 at 7:39 PM, Chris Dagdigian wrote: >> spew out a terabyte per day of raw data and many times that stuff needs to >> be post processed and distilled down into different forms. A nice little >> 8-core box running a shrink-wrap HPC product with a single support contact >> could find a nice little niche in non-datacenter areas where significant >> compute is needed nearby some other sort of dedicated instrument or device. > ... > On Wed, Apr 2, 2008 at 6:44 PM, Greg Byshenk wrote: >> a business will also have to find >> (and pay) someone to build and maintain the cluster. > > > Forgive me perhaps for being naive, but why can't a knowledgeable > teenager / college student be paid ~$10/hour plus on-call time to do a > setup like this? Presuming they only need to hire someone to do > setup/administration/support (and not the actual programming itself). Such teenager would both likely be available and capable of such things. A lot of businesses and bean-counters, however, will be leery of such an arrangement. Never mind some of the degree-minded faculty at our universities... Yeah, I could name names. Aside from that, the actual programming is a bit of a question. Once again, the teenager might be a good candidate to write the "code" since they're likely more current in coding that that tenured professor who's been mentoring grad students and writing proposals instead of his first love of research... A lot of folks in our place are using commercial "code" rather than writing their own. and then bending/folding their data to fit. And for that matter, (s)he can probably figure out how to automate mos tof the administration and monitornig! -- Gerry Creager -- gerry.creager@tamu.edu Texas Mesonet -- AATLT, Texas A&M University Cell: 979.229.5301 Office: 979.862.3982 FAX: 979.862.3983 Office: 1700 Research Parkway Ste 160, TAMU, College Station, TX 77843 From jmdavis1 at vcu.edu Thu Apr 3 13:50:30 2008 From: jmdavis1 at vcu.edu (Mike Davis) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <5a1205b30804031318p656b9d06hc55d891f95ab1b4d@mail.gmail.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> <06F64C9C-BA0E-4B77-910A-AB907D886A3E@myri.com> <5a1205b30804031318p656b9d06hc55d891f95ab1b4d@mail.gmail.com> Message-ID: <47F54316.5090508@vcu.edu> Kyle Spaans wrote: >> What if your data is sensitive or has HIPPA requirements? Do you want a >> part-timer having admin control of that data (regardless if it is Linux, >> Windows, or MacOSX)? >> > > Isn't that what NDAs are for? > _______________________________________________ > Beowulf mailing list, Beowulf@beowulf.org > To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf > NDA= lawsuit HIPPA combined with Sarb-Ox= massive fines or jail time. One does not want a high school student managing a cluster (or db machine or even laptop) that falls under HIPPA guidelines. My last 2 hires were internal. I began with more than 100 applicants in each case, narrowed to a shortlist of 5-7 for interviews. Discovered that at least 4 of that 7 who claimed linux or unix admin experience didn't even know how to create a new user. Their supposed batch experience (specified as SGE, PBS, Torque, LSF) was running jobs in the background from the command line. They had little experience with RAID and had not even bothered to do any research before showing up for the interview. In other words, they were unqualified. Mike Davis From hahn at mcmaster.ca Thu Apr 3 14:04:20 2008 From: hahn at mcmaster.ca (Mark Hahn) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <20080402224403.GY90095@gby1.aoes.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> Message-ID: > are a Windows shop, and thus the cost of hiring or training staff > would more than outweigh any savings on licenses. this is absurd! buying a windows cluster just means that the new cluster admins you need to hire or train will now labor under the dark-age infrastructure of windows rather than *nix. it's not as if any existing windows desktop or server admins will have mad mpi skilz... From mehmet.suzen at physics.org Thu Apr 3 12:48:06 2008 From: mehmet.suzen at physics.org (Mehmet Suzen) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <20080402224403.GY90095@gby1.aoes.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> Message-ID: <8c4169d00804031248s72aef8e4j670e5a3bc33eb433@mail.gmail.com> On Thu, Apr 3, 2008 at 12:44 AM, Greg Byshenk wrote: > On Wed, Apr 02, 2008 at 03:04:28PM -0700, Jon Forrest wrote: > > > But, the question remains. How can Microsoft compete with free? > > How much better will they have to be than standard Linux > > clusters before they get any mainstream interest? What technical > > features could they add that couldn't be added to a Linux > > cluster? > > The thing to remember is that a cluster (even one running "free" -- > as in beer -- software) is not without cost. Apart from hardware, > licenses (if required), etc., a business will also have to find > (and pay) someone to build and maintain the cluster. > It is VERY important to be clear about one point. This argument must NOT imply that using proprietary software to built a cluster does not require any work-force (someone) or no training needed to operate it. At the end of the day building a cluster is a technological business, you need to hire someone or a monkey in order to click on buttons for installation or maintenance. Having proprietary software is NOT a necessary and sufficient condition that professionals working on these products need less training then free software counter parts, ("free" as in freedom, NOT free beer software). > A former co-worker of mine now works elsewhere, at a company where > they are looking to build a cluster for some research he will be > doing, and it will almost certainly run Windows. And this is not > because there is something "better" about Windows, but because they > are a Windows shop, and thus the cost of hiring or training staff > would more than outweigh any savings on licenses. From mehmet.suzen at physics.org Thu Apr 3 13:18:26 2008 From: mehmet.suzen at physics.org (Mehmet Suzen) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <06F64C9C-BA0E-4B77-910A-AB907D886A3E@myri.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> <06F64C9C-BA0E-4B77-910A-AB907D886A3E@myri.com> Message-ID: <8c4169d00804031318g29c7b0a7n5b1c78f6c267d60a@mail.gmail.com> On Thu, Apr 3, 2008 at 10:13 PM, Scott Atchley wrote: > On Apr 3, 2008, at 3:52 PM, Kyle Spaans wrote: > > > On Wed, Apr 2, 2008 at 7:39 PM, Chris Dagdigian wrote: > > > > > spew out a terabyte per day of raw data and many times that stuff needs > to > > > be post processed and distilled down into different forms. A nice little > > > 8-core box running a shrink-wrap HPC product with a single support > contact > > > could find a nice little niche in non-datacenter areas where significant > > > compute is needed nearby some other sort of dedicated instrument or > device. > > > > > ... > > On Wed, Apr 2, 2008 at 6:44 PM, Greg Byshenk wrote: > > > > > a business will also have to find > > > (and pay) someone to build and maintain the cluster. > > > > > > > > > Forgive me perhaps for being naive, but why can't a knowledgeable > > teenager / college student be paid ~$10/hour plus on-call time to do a > > setup like this? Presuming they only need to hire someone to do > > setup/administration/support (and not the actual programming itself). > > > > What if your data is sensitive or has HIPPA requirements? Do you want a > part-timer having admin control of that data (regardless if it is Linux, > Windows, or MacOSX)? > Yes, part-time graduate student working on DoD project can do that for you. From mehmet.suzen at gmail.com Thu Apr 3 14:35:48 2008 From: mehmet.suzen at gmail.com (Mehmet Suzen) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <20080402224403.GY90095@gby1.aoes.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> Message-ID: <8c4169d00804031435l3703661dqaa652e8eacebacec@mail.gmail.com> On Thu, Apr 3, 2008 at 12:44 AM, Greg Byshenk wrote: > On Wed, Apr 02, 2008 at 03:04:28PM -0700, Jon Forrest wrote: > > > But, the question remains. How can Microsoft compete with free? > > How much better will they have to be than standard Linux > > clusters before they get any mainstream interest? What technical > > features could they add that couldn't be added to a Linux > > cluster? > > The thing to remember is that a cluster (even one running "free" -- > as in beer -- software) is not without cost. Apart from hardware, > licenses (if required), etc., a business will also have to find > (and pay) someone to build and maintain the cluster. > It is VERY important to be clear about one point. This argument must NOT imply that using proprietary software to built a cluster does not require any work-force (someone) or no training needed to operate it. At the end of the day building a cluster is a technological business, you need to hire someone or a monkey in order to click on buttons for installation or maintenance. Having proprietary software is NOT a necessary and sufficient condition that professionals working on these products need less training then free software counter parts, ("free" as in freedom, NOT free beer software). > A former co-worker of mine now works elsewhere, at a company where > they are looking to build a cluster for some research he will be > doing, and it will almost certainly run Windows. And this is not > because there is something "better" about Windows, but because they > are a Windows shop, and thus the cost of hiring or training staff > would more than outweigh any savings on licenses. > > > -- > Greg Byshenk - gbyshenk@byshenk.net - Leiden, NL > > > > _______________________________________________ > Beowulf mailing list, Beowulf@beowulf.org > To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf > From richard.walsh at comcast.net Thu Apr 3 15:07:09 2008 From: richard.walsh at comcast.net (richard.walsh@comcast.net) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? Message-ID: <040320082207.8819.47F5550D00063FC6000022732212059214089C040E99D20B9D0E080C079D@comcast.net> -------------- Original message -------------- From: Jon Forrest > That said, I just don't see how Microsoft's HPC server > can succeed. Yes, it seems impossible, but then what is success here? In the most limited sense this is a perimeter defending move by Microsoft. Now that a much larger part of the IT world is parallel and the Linux/HPC universe has a parallel cluster/savvy advantage along several dimensions. But also considered those new entrants into clustered server computing arena in proximity markets like finance where there is a lot of "application server" type work that is initiated from a Windows desktops. Then there is the improved workflow argument and the "high productivity computing" concept. That's mind share grabber with folks making spending decisions in these new arrival areas. Then there is raw staying power. If by success you mean total replacement and domination, not likely ... but if you mean capturing a somewhat larger share of the new arrivals into the market and protecting their flank from a end around stimulated by the parallel revolution ... then I would say, yes ... they can succeed. rbw -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.scyld.com/pipermail/beowulf/attachments/20080403/2f31277e/attachment.html From lindahl at pbm.com Thu Apr 3 21:08:12 2008 From: lindahl at pbm.com (Greg Lindahl) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] Re: SMPs + One processor machines = Heterogeneous Cluster In-Reply-To: References: Message-ID: <20080404040812.GA23797@bx9.net> On Thu, Apr 03, 2008 at 10:24:39AM +0800, Cally K wrote: > answers my question but just to be on the safe side, i would like to know > the response from the mailing list users. Well, just like the word "SMP", heterogeneous has a definition which depends on context. If I'm scheduling a job on a cluster, I think it's heterogeneous if some nodes have a different amount of memory or a different clock, or anything which makes their performance non-uniform. Most MPI codes only run as fast as their slowest process. That's the most picky definition, the least is the "do I have to recompile?" definition. -- greg From eugen at leitl.org Thu Apr 3 23:33:20 2008 From: eugen at leitl.org (Eugen Leitl) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> Message-ID: <20080404063320.GF9875@leitl.org> On Thu, Apr 03, 2008 at 05:04:20PM -0400, Mark Hahn wrote: > this is absurd! buying a windows cluster just means that the new > cluster admins you need to hire or train will now labor under the Do not underestimate the irrational fear of the command line. I see it every day at work. If it doesn't involve a rodent, it can't be done. > dark-age infrastructure of windows rather than *nix. it's not as if If you're a Windows shop, you don't know how bad it is. > any existing windows desktop or server admins will have mad mpi skilz... -- Eugen* Leitl leitl http://leitl.org ______________________________________________________________ ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org 8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE From kalpana0611 at gmail.com Thu Apr 3 23:43:47 2008 From: kalpana0611 at gmail.com (Cally K) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] Configuring mpich in a Pentium Dual Core Message-ID: Dear beowulf users I have an Intel pentium Dual Core machine, and I would like to know how to configure mpich on it, i used the smp option and is there any example programs that I can run besides the hello world that would tell me something like.... I am using both processors. I used this command ./configure --prefix=/home/kalpanak/Installation_Files/MPICH/mpichbuild ?with-device=ch_shmem ?with-comm=shared -rsh=ssh -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.scyld.com/pipermail/beowulf/attachments/20080404/d84ba760/attachment.html From csamuel at vpac.org Fri Apr 4 00:35:23 2008 From: csamuel at vpac.org (Chris Samuel) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <20080402224403.GY90095@gby1.aoes.com> Message-ID: <1327497775.282931207294523956.JavaMail.root@zimbra.vpac.org> ----- "Greg Byshenk" wrote: > A former co-worker of mine now works elsewhere, at a company where > they are looking to build a cluster for some research he will be > doing, and it will almost certainly run Windows. And this is not > because there is something "better" about Windows, but because they > are a Windows shop, and thus the cost of hiring or training staff > would more than outweigh any savings on licenses. This will also be a big factor for University ITS departments too which often seem to have (at least here in Australia) become MS only shops. -- Christopher Samuel - (03) 9925 4751 - Systems Manager The Victorian Partnership for Advanced Computing P.O. Box 201, Carlton South, VIC 3053, Australia VPAC is a not-for-profit Registered Research Agency From csamuel at vpac.org Fri Apr 4 00:47:09 2008 From: csamuel at vpac.org (Chris Samuel) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <566077032.282961207294969649.JavaMail.root@zimbra.vpac.org> Message-ID: <372505969.282981207295229776.JavaMail.root@zimbra.vpac.org> ----- "Jon Forrest" wrote: > What technical features could they add that couldn't > be added to a Linux cluster? Windows botnet clients & client access licenses ? :-) Seriously though, my concern is about the impact of the essential anti-virus, anti-malware and anti-spyware software on each node of the system be ? Who could seriously consider running *any* Windows box these days without them ? Not least on a system that is designed to have random people login in and run their applications on. cheers, Chris -- Christopher Samuel - (03) 9925 4751 - Systems Manager The Victorian Partnership for Advanced Computing P.O. Box 201, Carlton South, VIC 3053, Australia VPAC is a not-for-profit Registered Research Agency From john.hearns at streamline-computing.com Fri Apr 4 01:17:30 2008 From: john.hearns at streamline-computing.com (John Hearns) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] Configuring mpich in a Pentium Dual Core In-Reply-To: References: Message-ID: <47F5E41A.3020901@streamline-computing.com> Cally K wrote: > Dear beowulf users > > I have an Intel pentium Dual Core machine, and I would like to know how > to configure mpich on it, i used the smp option and is there any example > programs that I can run besides the hello world that would tell me > something like.... I am using both processors. The MPICH download comes with some simple example programs in C, C++ and Fortran. Try running couple of these. From geoff at galitz.org Fri Apr 4 03:19:42 2008 From: geoff at galitz.org (Geoff Galitz) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <372505969.282981207295229776.JavaMail.root@zimbra.vpac.org> Message-ID: <200804041020.m34AKeqq005122@b.mail.sonic.net> Just to be fair... well... maybe not fair... how about Devil's Advocate: Are any of the MS anti-malware apps more or less difficult to manage than their *NIX counterparts? Network layer firewalls, HIDS, file system integrity apps, spam containment and prevention apps and so on? Being a consultant these days means I no longer in live in fear of SSH, RPC/NFS or SNMP vulnerabilities all of which I have lost at least one cluster to in my career as a cluster admin... even with appropriate network level firewalling. More to the point of the thread: if we are talking about MS making greater inroads into the HPC market, then their most likely tactic is to convince the commercial app vendors to write and/or port their apps to the Windows platform. I've seen a number of animation and visualization apps running on MS clusters. They do exist and this is their best near-term chance of making those inroads. Perhaps Windows 7 with its anticipated modularization of Windows services will greatly improve their profile for running in a clustered environment. I don't think anyone expects currently established tools like NAMD (just as an example) to ever be ported over to Windows, but if MS can provide great development environments and support libraries for a new generation of scientists... well... maybe the future is less predictable than we may think. It seems HPC has been swinging away from performance and towards ease-of-use for some time, in my opinion. I've been doing clusters for approx 10 years, but like others I am now running Windows on my main workstation because it is the right combination of development environment, application availability and usability for me. Prior to Vista, this would not have been the case... so life and technology do evolve. Just my two cents. -geoff -----Original Message----- From: beowulf-bounces@beowulf.org [mailto:beowulf-bounces@beowulf.org] On Behalf Of Chris Samuel Sent: Freitag, 4. April 2008 09:47 To: Beowulf Mailing List Subject: Re: [Beowulf] How Can Microsoft's HPC Server Succeed? ----- "Jon Forrest" wrote: > What technical features could they add that couldn't > be added to a Linux cluster? Windows botnet clients & client access licenses ? :-) Seriously though, my concern is about the impact of the essential anti-virus, anti-malware and anti-spyware software on each node of the system be ? Who could seriously consider running *any* Windows box these days without them ? Not least on a system that is designed to have random people login in and run their applications on. cheers, Chris -- Christopher Samuel - (03) 9925 4751 - Systems Manager The Victorian Partnership for Advanced Computing P.O. Box 201, Carlton South, VIC 3053, Australia VPAC is a not-for-profit Registered Research Agency _______________________________________________ Beowulf mailing list, Beowulf@beowulf.org To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf From mark.kosmowski at gmail.com Fri Apr 4 04:49:16 2008 From: mark.kosmowski at gmail.com (Mark Kosmowski) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] cluster admin employment Message-ID: > Message: 7 > Date: Thu, 03 Apr 2008 16:50:30 -0400 > From: Mike Davis > Subject: Re: [Beowulf] How Can Microsoft's HPC Server Succeed? > To: Kyle Spaans <3lucid@gmail.com> > Cc: Beowulf List > Message-ID: <47F54316.5090508@vcu.edu> > Content-Type: text/plain; charset=UTF-8; format=flowed > > Kyle Spaans wrote: > >> What if your data is sensitive or has HIPPA requirements? Do you want a > >> part-timer having admin control of that data (regardless if it is Linux, > >> Windows, or MacOSX)? > >> > > > > Isn't that what NDAs are for? > > _______________________________________________ > > Beowulf mailing list, Beowulf@beowulf.org > > To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf > > > > > NDA= lawsuit > > HIPPA combined with Sarb-Ox= massive fines or jail time. > > > One does not want a high school student managing a cluster (or db > machine or even laptop) that falls under HIPPA guidelines. > > > My last 2 hires were internal. I began with more than 100 applicants in > each case, narrowed to a shortlist of 5-7 for interviews. Discovered > that at least 4 of that 7 who claimed linux or unix admin experience > didn't even know how to create a new user. Their supposed batch > experience (specified as SGE, PBS, Torque, LSF) was running jobs in the > background from the command line. They had little experience with RAID > and had not even bothered to do any research before showing up for the > interview. In other words, they were unqualified. > > Mike Davis > For Mike and the rest of you: If one of these unqualified applicants had accurately mapped out their deficiencies and let you know that they would need a little training, would said applicant have been considered at all or would their resume have been filed in the "circular file" (slang for the trash bin)? Does someone who has maintained a personal cluster for their graduate research have any reasonable chance of landing some sort of admin position? Given that state of the economy, I've been kind of hoping that I could look into admin positions as a backup plan if a position in my field is not readily available once I finish my doctoral studies - it would be helpful to know whether this is a realistic possibility. Thanks, Mark Kosmowski From landman at scalableinformatics.com Fri Apr 4 05:26:07 2008 From: landman at scalableinformatics.com (Joe Landman) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <1327497775.282931207294523956.JavaMail.root@zimbra.vpac.org> References: <1327497775.282931207294523956.JavaMail.root@zimbra.vpac.org> Message-ID: <47F61E5F.7030009@scalableinformatics.com> Chris Samuel wrote: > ----- "Greg Byshenk" wrote: > >> A former co-worker of mine now works elsewhere, at a company where >> they are looking to build a cluster for some research he will be >> doing, and it will almost certainly run Windows. And this is not >> because there is something "better" about Windows, but because they >> are a Windows shop, and thus the cost of hiring or training staff >> would more than outweigh any savings on licenses. > > This will also be a big factor for University ITS departments > too which often seem to have (at least here in Australia) become > MS only shops. Interesting ... in the commercial space, I am seeing something different. Noticed it starting in earnest sometime last year. Some of the larger shops (and many of the smaller ones) have been shifting to Linux desktops. They use their laptops for office things, though an increasing number of them are using linux on their laptops. -- Joseph Landman, Ph.D Founder and CEO Scalable Informatics LLC, email: landman@scalableinformatics.com web : http://www.scalableinformatics.com http://jackrabbit.scalableinformatics.com phone: +1 734 786 8423 fax : +1 866 888 3112 cell : +1 734 612 4615 From rgb at phy.duke.edu Fri Apr 4 05:22:08 2008 From: rgb at phy.duke.edu (Robert G. Brown) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] How Can Microsoft's HPC Server Succeed? In-Reply-To: <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> References: <47F402EC.3050701@berkeley.edu> <20080402224403.GY90095@gby1.aoes.com> <5a1205b30804031252t1b96aa27oa0c1d70e821582bd@mail.gmail.com> Message-ID: On Thu, 3 Apr 2008, Kyle Spaans wrote: > On Wed, Apr 2, 2008 at 7:39 PM, Chris Dagdigian wrote: >> spew out a terabyte per day of raw data and many times that stuff needs to >> be post processed and distilled down into different forms. A nice little >> 8-core box running a shrink-wrap HPC product with a single support contact >> could find a nice little niche in non-datacenter areas where significant >> compute is needed nearby some other sort of dedicated instrument or device. > ... > On Wed, Apr 2, 2008 at 6:44 PM, Greg Byshenk wrote: >> a business will also have to find >> (and pay) someone to build and maintain the cluster. > > > Forgive me perhaps for being naive, but why can't a knowledgeable > teenager / college student be paid ~$10/hour plus on-call time to do a > setup like this? Presuming they only need to hire someone to do > setup/administration/support (and not the actual programming itself). For parts of the beowulf "revolution", this is close to what happened -- opportunity cost time, no professional support. However, a) That doesn't work for corporations or big research computing groups at Universities or for University research programs where the primary researchers aren't computer geeks; and b) You are off by an order of magnitude in commercial rates for this sort of professional support. Hourly rates range from a barebones minimum of perhaps $20/hour plus benefits for somebody with skills at the level you describe (who simply wouldn't do for anything but a tiny cluster in a non-complex environment) to $75-200/hour for offsite pro-grade management and/or consulting. Onsite professionals make from $50K/year (for a fairly young, not yet terribly experienced) to perhaps $80K/year to do actual hands-on sysadmin and network engineering in a University, with the range being (correct me if I'm wrong, people) $75K to $150K/year for corporate, with the upper bounds of both ranges held by people who are also managers and who run junior sysadmins in a group of some sort. Offsite is much more expensive, STARTING at hourly rates corresponding to the $150K/year for a really good full time employee. Consultants are typically real experts, although there as always there is considerable variation, caveat emptor etc. rgb > _______________________________________________ > Beowulf mailing list, Beowulf@beowulf.org > To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf > -- Robert G. Brown Phone(cell): 1-919-280-8443 Duke University Physics Dept, Box 90305 Durham, N.C. 27708-0305 Web: http://www.phy.duke.edu/~rgb Book of Lilith Website: http://www.phy.duke.edu/~rgb/Lilith/Lilith.php Lulu Bookstore: http://stores.lulu.com/store.php?fAcctID=877977 From landman at scalableinformatics.com Fri Apr 4 05:42:05 2008 From: landman at scalableinformatics.com (Joe Landman) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] cluster admin employment In-Reply-To: References: Message-ID: <47F6221D.50602@scalableinformatics.com> Mark Kosmowski wrote: > For Mike and the rest of you: > > If one of these unqualified applicants had accurately mapped out their > deficiencies and let you know that they would need a little training, > would said applicant have been considered at all or would their resume > have been filed in the "circular file" (slang for the trash bin)? Hi Mark: As an employer, I'll provide my perspective. I want people who I can assign a task to and a time line, and have that task delivered in that time line. If the person doesn't know something they need for the task, I expect them to be smart enough to recognize it, intelligent enough to figure out what they need to learn, and then learn it. All within our time and deliverable constraints. Even better, someone for whom I can define a nebulous goal, a due date, and then let them tear into it, figuring out all the details on their own. FWIW: I am not convinced that certifications map into this in any reasonable way, nor am I convinced that they correlate with the ability to learn and execute in a limited time window based upon what you learn. My experience (dealing with certified people) has been quite the opposite. > Does someone who has maintained a personal cluster for their graduate > research have any reasonable chance of landing some sort of admin > position? Given that state of the economy, I've been kind of hoping > that I could look into admin positions as a backup plan if a position > in my field is not readily available once I finish my doctoral studies > - it would be helpful to know whether this is a realistic possibility. I would aim higher than admin. A doctoral degree implies one has the critical thinking and reasoning skills well honed for identifying, understanding, and solving problems (caveat: see my point on certifications above). Do a realistic self inventory. What items would an employer find valuable? During grad school, I met students with perfect GPAs, who could regurgitate qual-exam problems and their solutions, but were lost on basic reasoning skills. And I met students who, once a problem got stuck in their craw, would keep hammering on it until the "got it done". Well, some of them would be motivated to get it done in a reasonable time. They would learn what they needed to get the job done. Guess which group is interesting to me as an employer. Joe -- Joseph Landman, Ph.D Founder and CEO Scalable Informatics LLC, email: landman@scalableinformatics.com web : http://www.scalableinformatics.com http://jackrabbit.scalableinformatics.com phone: +1 734 786 8423 fax : +1 866 888 3112 cell : +1 734 612 4615 From jmdavis1 at vcu.edu Fri Apr 4 05:49:47 2008 From: jmdavis1 at vcu.edu (Mike Davis) Date: Sat Jul 19 01:07:00 2008 Subject: [Beowulf] cluster admin employment In-Reply-To: References: Message-ID: <47F623EB.7090805@vcu.edu> Mark Kosmowski wrote: > For Mike and the rest of you: > > If one of these unqualified applican