[Beowulf] Erlang as a language for Beowulf applications
H.Vidal, Jr.
hvidal at tesseract-tech.com
Thu Sep 6 10:03:50 PDT 2007
Thanks for the heads-up. Yes, I noticed this review just
as I was finishing the very same book.
It seems that Erlang is not best suited for HPC 'as is'
but could, conceivably, be used as a very applicable
infrastructure for HPC. One might envision a system where
the hard lifting is done via C, under the guidance of Erlang
processes.
The other thing that is potentially quite nice is the use
of Erlang as a kind of fault-tolerant batch control system;
since it has the notion of failovers built in, a common
problem of 'dead nodes' (whether software crash or hardware
failure) in the course of very long compute jobs might be
addressed via Erlang's ability to build 'warm start' processing
into the application level.
Certainly Erlang, multi-cores, Intel Thread Building Blocks,
and other interesting topics in parallel software and hardware
make for interesting times. I am quite involved in working with
advanced (and rather gifted) high school students in their
computer science studies, and one of the things I am championing
is the notion that parallel programming and related engineering
should be part of the core curriculum for these students as they
represent the next generation of engineers and scientists that
will inherit these sorts of applications and problems on a
day-to-day basis in their future work. Again, interesting times.
Thanks again.
hv
Peter St. John wrote:
> HV,
> Slahdot just posted a review of a recent book on Ehrlang,
> http://books.slashdot.org/article.pl?sid=07/09/05/1410213
> Peter
>
>
> On 9/3/07, *H.Vidal, Jr.* <hvidal at tesseract-tech.com
> <mailto:hvidal at tesseract-tech.com>> wrote:
>
> Hello.
>
> I have been exploring a range of technologies for parallel
> applications, some production-level, some experimental.
>
> I am curious if anyone on this list has done any work
> with the language Erlang and/or considers it viable
> for scientific apps. It seems to be quite mature, has
> well developed 'process' based semantics with intrinsic message
> passing, is light-weight for multi-process creation,
> support application-level fault tolerance (quite applicable
> for failures in long computations...)
> and is production level, though not well known in the US.
>
> Any comments?
>
> hv
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org <mailto:Beowulf at beowulf.org>
> To change your subscription (digest mode or unsubscribe) visit
> http://www.beowulf.org/mailman/listinfo/beowulf
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
More information about the Beowulf
mailing list