(trans)portable Beowulf - (slightly O/T)

Victor Karyo vkaryo at hotmail.com
Wed Nov 15 12:13:07 PST 2000

Found this on a site (www.sciencewise.com).
Seems the DoD could use portable beowulfs.
(sorry for the formating - or lack of).



Department of the Air Force, Air Force Materiel Command, AFRL -- Rome
Research Site, AFRL/Information Directorate, 26 Electronic Parkway,
Rome, NY, 13441-4514

EXPLOITATION TOOLS SOL Reference-Number-BAA-00-01-OFKPA POC Joetta Bernhard,
Contracting Officer, Phone (315) 330-2308, Fax (315) 330-7790, Email
bernhard at rl.af.mil. WEB: Visit this URL for the latest information about
. E-MAIL: Joetta Bernhard,
bernhard at rl.af.mil. Contact: AFRL Program Manager: Jon Jones, e-mail
jon.jones at rl.af.mil, 315/330-1665; Technical Point of Contact: Richard
Gassner, e-mail richard.gassner at rl.af.mil, 315/330-3574; Mark
Pronobis, e-mail mark.pronobis at rl.af.mil, 315/330-3841 and Steve Scott,
e-mail steve.scott at rl.af.mil, 315/330-4414; Contracting Officer, Joetta
Bernhard, e-mail joetta.bernhard at rl.af.mil, 315/330-2308. The Air Force
Research Laboratory Information Directorate, Information and
Intelligence Exploitation Division is seeking innovative solutions for
the development of automated tools to exploit MTI data and incorporate
information fusion from current and future sensor systems. This will
offer the warfighter an enhanced operational capability to
automatically exploit MTI data in order to locate, identify, and track
high value ground moving targets. Current MTI exploitation techniques
are manual, operator intensive, and very limited. This effort will
also provide algorithms to enable the optimal fusion of resources for
the timely detection, track, and identification of ground moving
targets. There are currently more targets than resources in any concept
of operation so that adequate optimization of resources is critical to
being able to engage moving vehicles. This MTI exploitation BAA will
focus on the following technology areas: ground moving target tracking,
motion pattern analysis, behavioral pattern analysis, geo-registration,
feature aided tracking, multiple platform tracking, and resource
allocation and scheduling. In the area of ground moving target
tracking, innovative algorithms are sought to allow current
surveillance platforms to automatically track targets in a variety of
background environments. Motion pattern analysis tools may also be
developed to analyze MTI tracks and identify entities or groups of
entities that may be of military significance. Examples of this include
the detection of convoys and formations, determination of sources and
sinks via traffic flow analysis, and determination of lines of
communication. Behavioral pattern analysis tools are needed to
determine the intent of entities or groups of entities based on
terrain, doctrine, and historical data. An example would be the
determination of a probable hide site for a missile launcher based on
measurements indicating the launch site. Geo-registration is needed to
accurately determine target location by reducing tacking errors from
different sensors, specifically registering RF tags and images from
multiple platforms. Feature aided tracking may be developed to track
targets based on certain features, some of which include velocity,
vehicle type (track vs. wheels), HRR, and ISAR. Multiple platform
tracking will improve target recognition and location by utilizing MTI
data gathered from various sensors. The combination of these
technologies will provide a powerful tool set for operators to exploit
MTI data and will enable the creation of a more comprehensive picture
of the battlespace. Finally, in the area of resource allocation the
focus will be that of moving target indication and the scheduling
required to detect, track, and exploit moving targets whether the
target be airborne or ground based. The scheduling algorithms will be
capable of being integrated into the Air Forces Time Critical Targeting
Cell for the real time allocation of resources to assist in the
detection, track, and identification of time sensitive targets. The
scheduling algorithms may also be integrated with the targeting
assignment toolbox, namely the Joint Targeting Toolbox. Information
fusion will require the gathering and processing of information from
various sensors, platforms, and multiple systems to include MTI fusion
with HRR reports, imagery, and SIGINT. Advanced concepts for
information fusion technology, which includes multilevel
sensor/platform/system resource management, central/distributed fusion
management, and database information management, also need to be
developed to achieve this goal. Multilevel sensor/platform/system
resource management is needed to provide a seamless environment for the
user. This involves global resource management where platform cueing
and fusion of various sensors, platforms, and/or integration of
multiple systems will provide extensive data. Central/distributed
fusion management may include optimizing data distinctiveness, defining
mechanisms for information push/pull, and defining levels of
responsibility. Database information management needs to be enhanced by
concepts of robustness such as model-based reasoning, context
sensitivity and efficiency. The database support to the algorithms
needs to be upgraded by concepts such as client/server relationships
that take advantage of object oriented software. New parallel
techniques must be explored to allow implementation of algorithms for
real time use. This effort seeks to leverage existing technologies such
as the Ground Moving Target Indication (GMTI) Multi-Platform fusion
algorithms and develop a resource scheduling algorithm analyzing
terrain blockage, cross cueing capabilities, revisit rates, and
detection accuracy. The Air Force currently can schedule sensor
resources such as Joint STARS and Global Hawk in real time using
dedicated communication links between the ground stations and the
platforms. Currently no cross cueing is being performed optimizing
coverage and the information flow between platforms. The same goes for
space assets as well, where timeliness of the data is more of an issue
because of the collection times associated with space assets. However,
space assets provide a new foundation in that they provide consistent
coverage of hostile target areas over a continuous 24 hour cycle. This
provides for normalization to be developed over the movement patterns
associated with normal operations. The effort will leverage existing
developments in the operational area such as the Time Critical
Targeting Aid and Moving Target Indicator Exploitation (MTIX) and
develop on these architectures making it easily adaptable to existing
architectures for technology transition. From all aspects of
information fusion technology, measures of performance and measures of
effectiveness need to be developed and integrated into these new
concepts. This will provide methods for consistent world views. THIS
FORMAL PROPOSAL AT THIS TIME. Offerers are required to submit an
original and three (3) copies of a ten (10) page or less white paper
with a cover letter indicating whether the offeror is a large, small,
women-owned small, or small disadvantaged business, or Historically
Black College, University, or Minority Institution. Small business
offerors must also send one (1) copy of the cover letter only, by First
Class mail (DO NOT SEND CLASSIFIED), to AFRL/IFB, ATTN: Janis Norelli,
Director of Small Business, 26 Electronic Parkway; Rome NY 13441-4514.
The white paper will be formatted as follows: Section A: Title, Period
of Performance, Cost of Task, Name of Company; Section B: Task
Objective; and Section C: Technical Summary. All responses to this
announcement must be addressed to AFRL/IFEA, ATTN: Rick Gassner,
Reference BAA 00-01-IFKPA, 32 Brooks Road, Rome NY 13441-4114. Multiple
white papers within the purview of this announcement may be submitted
by each offeror. White papers are due for FY00 not later than 15
December 1999. It is recommended that white papers be received by the
following dates to maximize the possibility of award: FY01-30Jul00;
FY02-30Jul01; FY03-30Jul02; FY04-30Jul03. White papers will be accepted
at any time, but it is less likely that funding will be available in
each Fiscal Year after the dates cited. The purpose of the white paper
is to preclude unwarranted effort on the part of an offeror whose
proposed work is not of interest to the Government. Those white papers
found to be consistent with the intent of this BAA may be invited to
submit a technical and cost proposal. Such invitation does not assure
that the submitting organization will be awarded a contract. Complete
instructions for proposal preparation will be forwarded with the
invitation for proposal submission. Evaluation of proposals will be
performed using the following criteria: a) The overall scientific
and/or technical merits of the proposal, b) The innovativeness of the
proposed approach and/or techniques, and c) The reasonableness and
realism of proposed costs and fees. In addition, the Government will
consider other factors, such as past and present performance on recent
Government contracts and the capacity and capability to perform the
requirements of the effort. The criteria used to evaluate proposals
will also be used to determine whether white papers submitted are
consistent with the intent of this BAA and of interest to the
Government. No further evaluation criteria will be used in selecting
the proposals. Individual proposal evaluations will be based on
acceptability or unacceptability without regard to other proposals
submitted under this BAA. Options are discouraged and unpriced options
will not be considered for award. Principle funding of this BAA and
anticipated awards will start approximately FY00. Individual awards
will not normally exceed 24 months in duration, with dollar amounts
normally ranging from $100K to $1.5M. Total funding for this BAA is
$24.0M. Awards of efforts as a result of this BAA will be in the form
of contracts, grants, cooperative agreements, or other transactions
depending upon the nature of the work proposed. Foreign-owned offerors
are advised that they are excluded from participating at the prime
contract level. Foreign or foreign-owned offerers should immediately
contact the contracting focal point, Joetta A. Bernhard, for
information if they contemplate responding. Data subject to export
control constraints may be involved and only firms on the Certified
Contract Access List (CCAL) will be allowed access to such data. The
cost of preparing proposals in response to this announcement is not an
allowable expense to the normal bid and proposal indirect cost in AFR
31.205-18. Depending on the work to be performed, the offeror may
require a SECRET facility clearance and safeguarding capability;
therefore, personnel identified for assignment to a classified effort
must be cleared for access to SECRET information at a time of award. An
Ombudsman has been appointed to hear significant concerns from offerors
or potential offerors during the proposal development phase of this
acquisition. Routine questions are not considered to be "significant
concerns" and should be communicated directly to the Contracting
Officer, Joetta A. Bernhard, (315)330-2308. The purpose of the
Ombudsman is not to diminish the authority of the Contracting Officer
or Program Manager, but to communicate contractor concerns, issues,
disagreements and recommendations to the appropriate Government
personnel. The Ombudsman for this acquisition is Vincent Palmiero,
Deputy Chief, AFRL/IFK at (315) 330-7746. When requested, the Ombudsman
will maintain strict confidentiality as to the source of the concern.
The Ombudsman does not participate in the evaluation of the proposals
or in the source selection process. To receive a copy of the AFRL "BAA
& PRDA: A Guide for Industry," Sep 1996 (Rev), write to AFRL/IFK,
ATTN: Lucille Argenzia, 26 Electronics Parkway, Rome, NY 13441-4514, or
fax her at FAX 315-330-4728, or the guide may be accessed at:
http://www.if.afrl.af.mil/div/IFK/bp-guide.html. This BAA is open and
effective until 30 Dec 04 All responsible firms may submit a white
paper, which shall be considered. Respondents are asked to provide
their Commercial and Government Entity (CAGE) number with their
submission and reference BAA 00-01-IFKPA. Only Contracting Officers are
legally authorized to commit the Posted 11/18/99 (D-SN401833). (0322)

Keycode Keyword
A000610 Air Force Office of Scientific Research
1015530 Ground Stations
101500A All Space Sciences
1000000 Science & Technology & Mathematics & Computer Science
0505027 Minority Institution
1010041 Optimization
0000034 Research
1016000 Technology
101600A All Technology
1015000 Space Sciences
1004015 Artificial Intelligence & Cybernetics
100400A All Computer Science
1004000 Computer Science
110400A All Military Sciences & History
1104000 Military Sciences & History
1016010 Acoustic Target Detection
1004173 Algorithms
1004066 Parallel Algorithms
0607021 Surveillance Systems
1015127 Airborne & Spaceborne Computers
1004063 Distributed Systems
1004047 Real-Time Computing
1004030 Computer Architecture
0505023 Hispanic-Serving Institutions
0505011 Historically Black Colleges & Universities

----Original Message Follows----
From: "Robert G. Brown" <rgb at phy.duke.edu>
To: Ken <lowther at att.net>
CC: Jim Lux <James.P.Lux at jpl.nasa.gov>, beowulf at beowulf.org
Subject: Re: (trans)portable Beowulf
Date: Tue, 14 Nov 2000 09:01:27 -0500 (EST)

On Tue, 14 Nov 2000, Ken wrote:

 > Jim Lux wrote:
 > >
 > >
 > > Real power consumption - how much current on each voltage is being 
drawn by
 > > a typical motherboard?
 > My whole system, - monitor of course, draws about .77 amps nominal, and
 > .97 amps with both processors fully loaded. (Dual 466 Celeron's) I
 > didn't try pulling the rest of the components.  I wasn't interested in
 > that at the time I checked.

I think that this is fairly typical.  This has been discussed before --
do a search on the list archives.  I think that "most" systems consume
70-100 watts (which is the important measure, not current per se, as
you'll be providing the current at a different voltage).  However, it
varies depending on just what you have on your system.  A box with four
hard drives, max memory, a CD, a floppy, a big/fast video card, and a
couple or three NICs will consume a bit more than a bare ATX motherboard
with a single PXE NIC, a CPU, and enough memory to run.

You don't mention budget (that I recall) in your project description.
If "price is no object" you might look carefully at e.g. laptop-based
components.  There are matched CPU, motherboard, memory sets out there
that consume far less power than an ATX motherboard and OTC CPUs.  I
think I recall IBM just announcing something that will allow their
newest laptops to run 8 hours active on a single charge.

I'm certain that these systems are more expensive (by a factor of 2-4,
maybe?) than a straight ATX-based solution, but you might try talking
IBM into "participating" with you on the project and selling you e.g. 8
filled motherboards for a nominal price -- it would be great press for
them if you were to build an 8 CPU beowulf that would fit into the
volume of a carryon bag and run 24 hours on a car battery.  You might
even convince them to make it the prototype of a new product -- stick a
folddown flatpanel monitor and keyboard on top of it with a switch, put
a little 8 port 100Base switch inside and voila!  IBM's new "field
supercomputer".  The military would probably lap it up.

Also remember that with Scyld, the nodes can be REALLY thin -- as you
note, PXE NIC, CPU, memory.  I don't even see the point of a video
interface, once you have the nodes configured to boot via the NIC and
without video or keyboard (which might require a single trip through the
BIOS).  Once configured, they boot.  Once booted, they run.  If there
are problems that survive a power cycle, they are almost certainly
hardware.  So carry spare parts and a single cheap video card that you
can move around to debug hardware problems, maybe.

Remember that every additional entity (video, sound, whatever) on the
node motherboards consumes power, and keeping power down BOTH lets you
run longer AND lets you run cooler.  A desktop ATX-motherboard based
system is likely to draw at least 400 watts on 8 nodes, and car
batteries (or any other kind of portable battery that weighs less than 1
pound/watt provided) get tired real fast at this sort of draw level.  I
really think that you're going to HAVE to look for very low power
solutions and get your 8 node consumption down in the 100-200 watt level
or even less or you'll only be able to run for around an hour or two
absolute max with the car turned off or no generator running.

Another company to check out is Transmeta (Linus Torvald's employer).
Their "Crusoe" processor is featured in e.g. the Sony VAIO C1 ultralight
notebook.  The TM processors are designed to have extremely low power
draw and internal idle modes to conserve power even further.  You could
actually probably build your whole beowulf out of a stack of VAIO C1's,
in which case the finished 8 node unit would likely weigh about 22
pounds and would run for about 5 hours in the field (presuming you can
find or cobble together a low power 9 port switch and a battery to run
it for 5 hours standalone).  Of course, it would probably cost a lot
compared to a conventional beowulf, but then, one cannot carry a
conventional beowulf -- and your lunch -- to the top of a volcano or out
into the desert in a single backpack and use it there for half a day.

Be very careful -- I know of no benchmarks for the TM CPUs and so I have
no idea how their clock translates into e.g. MFLOPS in a HPC
application.  You would almost certainly want to benchmark them before
considering this.

Again, talk to Transmeta and see if they are willing to "participate"
and make hardware available to you at cost for the publicity.  If you
could get the raw motherboards, CPUs, PCMCIA interface and memory and
could leave off the monitors, floppies, keyboards, sound and so forth
you could probably lighten the load and reduce power.  Even if you can
only get the notebooks, if you are willing to do a little mechanical
surgery you can probably strip off the folding display and keyboards
(leaving plugs in place so you can use just one for any node if you need
to).  That might get your weight down to what, 15 pounds?  Heck, my 15"
laptop weighs more than half of that.

Again, I'd guess that the army would just love a 10 pound supercomputer
that could be attached to e.g. artillery pieces along with a GPS, an
anemometer, a compass, a level and a gyroscope for inputs, be fed
absolute coordinates, and proceed to calculate precise firing angles in
real time correcting for things like latitude, absolute distance from
the earth's axis of firing point and target point and so forth (e.g.
coriolis forces).  Oh, they may already have this sort of thing but
then, they may not.  The Star Wars folks will also NEED stuff like this
and beyond if they are to have any hope of doing realtime corrections to
inbound ballistic or actively redirected trajectories.  You could likely
put in a proposal and get them to fund the whole thing lavishly.

Good luck.  Sounds like an interesting idea.  Please report to the list
when you get it built and consider writing a fully descriptive "article"
to be included as a section in the Beowulf book Doug Eadline and I are
writing (and openly providing) online.  We guarantee authorial
recognition only, although who knows -- it might make money some day.


Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu

Beowulf mailing list
Beowulf at beowulf.org

Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.

Share information about yourself, create your own public profile at 

More information about the Beowulf mailing list