home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.parallel
- Path: sparky!uunet!cs.utexas.edu!sdd.hp.com!swrinde!gatech!hubcap!fpst
- From: cfreese@super.super.org (Craig F. Reese)
- Subject: Re: 16-node hypercube
- Message-ID: <1992Nov23.200917.15963@hubcap.clemson.edu>
- Apparently-To: uunet!comp-parallel
- Summary: questions
- Sender: cfreese@super.super.org
- Nntp-Posting-Host: super
- Organization: Supercomputing Research Center (Bowie, MD)
- References: <1992Nov23.125706.8641@hubcap.clemson.edu>
- Distribution: na
- Date: Mon, 23 Nov 1992 17:50:37 GMT
- Approved: parallel@hubcap.clemson.edu
- Lines: 54
-
- In article <1992Nov23.125706.8641@hubcap.clemson.edu> black@studsys.mscs.mu.edu (benjamin j black ) writes:
- >
- >Keywords:speed, speed, and more speed
- >
- > 16 DSP CPU's (50MFLOPS each)
- > 2 MB 20ns SRAM per processor
- > 2 GB Disk space (SCSI-2)
- >
- > SPACE sub-microkernel OS, Dataparallel C compiler,etc.
- >
- > Any questions?
- > black@studsys.mscs.mu.edu
- >
- As a matter of fact, yes.
-
- Which DSP?
- How many wait states on the external memory?
- What are the hypercube links? (i.e. are they C40 comm links
- or something home grown. If so what is the performance.)
- What is the limit on scalability?
- Is the interconnet reconfigurable to say make a shuffle-exchange,
- ring, mesh, etc... (i.e. can I recable the thing?)
- Any shared memory among the processors?
- Is the disk space distributed among the processors or from a
- single node/frontend.
- What facilities are there for parallel I/O.
- Is the machine standalone or does it have a frontend. If
- so what is it?
- What applications have actually been developed, tested, and
- benchmarked?
-
- And lastly...
-
- From where can I FTP some PostScript technical reports on the
- hardware and software.
-
- Craig
-
- *** The opinions expressed are my own and do not necessarily reflect
- *** those of any other land dwelling mammals or my management....
-
- "The problem ain't what we don't know; it's what we know that just ain't so.
- Either we take familiar things so much for granted that we never think about
- how they originated, or we "know" too much about them to investigate closely."
- -----------------
- Craig F. Reese Email: cfreese@super.org
- Institute for Defense Analyses/
- Supercomputing Research Center
- 17100 Science Dr.
- Bowie, MD 20715-4300
-
-
-
-
-