cpmwc@marlin.jcu.edu.au (Matthew W Crowd) (06/03/91)
Not that I'm going to buy one, but how much does the Cray cost ??
What are the system configurations and hardware specs ?
matt.
--
=======================================================================
Matt Crowd cpmwc@marlin.jcu.edu.au | Thinking of a
| Master Plan.....
Resident Computer Science FTP Mutant | Eric.Bdre@sequoia.cray.com (Dennis Ellis) (06/04/91)
In article <1991Jun3.034841.25633@marlin.jcu.edu.au> cpmwc@marlin.jcu.edu.au (Matthew W Crowd) writes: > >Not that I'm going to buy one, but how much does the Cray cost ?? Why not? We could sign you up for one right now, if you want. > >What are the system configurations and hardware specs ? The system configurations are (roughly): XMS about $300,000 YMP-EL about $400,000 to $1,000,000 YMP2E about $2,000,000 to 5,000,000 YMP4E about $4,000,000 to 10,000,000 YMP8I about $15,000,000 YMP8E about $20,000,000 These are ranges (obviously) and approximate prices (depending on the details of the configurations and add-ons) For more details or more accurate pricing, see your local Cray salesperson. But this should give you some idea of where the various models fall in the pricing scheme. -- ******************************************************************************* * Dennis Ellis (dre@sequoia.cray.com) * * * Cray Research, Inc. * * * Eagan, MN 55121 * All the normal stuff about opinions.. *
mccalpin@perelandra.cms.udel.edu (John D. McCalpin) (06/04/91)
>>>>> On 3 Jun 91 17:16:15 GMT, dre@sequoia.cray.com (Dennis Ellis) said:
Dennis> The system configurations are (roughly):
Dennis> XMS about $300,000
Dennis> YMP-EL about $400,000 to $1,000,000
Dennis> YMP2E about $2,000,000 to 5,000,000
Dennis> YMP4E about $4,000,000 to 10,000,000
Dennis> YMP8I about $15,000,000
Dennis> YMP8E about $20,000,000
Dennis> These are ranges (obviously) and approximate prices
Dennis> (depending on the details of the configurations and add-ons)
I have never seen a technical summary of the models XMS or Y/MP-EL.
What are their vital statistics?
If the XMS is really X/MP compatible, then I guess all I need is the
clock speed. If not, what is the maximum sustainable memory bandwidth
from main memory to vector registers?
Ditto for the YMP/EL.
Thanks!
--
John D. McCalpin mccalpin@perelandra.cms.udel.edu
Assistant Professor mccalpin@brahms.udel.edu
College of Marine Studies, U. Del. J.MCCALPIN/OMNETslamont@network.ucsd.edu (Steve Lamont) (06/04/91)
In article <MCCALPIN.91Jun3142209@pereland.cms.udel.edu> mccalpin@perelandra.cms.udel.edu (John D. McCalpin) writes: >I have never seen a technical summary of the models XMS or Y/MP-EL. > >What are their vital statistics? > >If the XMS is really X/MP compatible, then I guess all I need is the >clock speed. If not, what is the maximum sustainable memory bandwidth >from main memory to vector registers? The XMS is the Supertek S-1 in Cray clothing. 40 nSec clock. Somewhere around a quarter of a single X/MP head. >Ditto for the YMP/EL. Dunno about that one -- I had notes from the dog and pony CRI put on a few weeks ago but can't seem to find them. About 1/4 of a Y/MP rings a bell but I could be making that up. Cray-ons? spl (the p stands for probably comes with a hand crank instead of a bench seat...) -- Steve Lamont, SciViGuy -- (408) 646-2752 -- a guest at network.ucsd.edu -- NPS Confuser Center / Code 51 / Naval Postgraduate School / Monterey, CA 93943 "When people are programming virtual 5-D webs of glowing spidersilk by pure thought power -- there will still be hackers." T.Neff in alt.folklore.computers
dre@sequoia.cray.com (Dennis Ellis) (06/05/91)
In article <MCCALPIN.91Jun3142209@pereland.cms.udel.edu> mccalpin@perelandra.cms.udel.edu (John D. McCalpin) writes: > >I have never seen a technical summary of the models XMS or Y/MP-EL. > >What are their vital statistics? > >If the XMS is really X/MP compatible, then I guess all I need is the >clock speed. If not, what is the maximum sustainable memory bandwidth >from main memory to vector registers? > The XMS is X-MP compatable. It runs UNICOS and all the applications optimized for the X-MP. The general characteristics are: 55 ns clock 64-bit architecture 64 or 128 MB memory 80 ns DRAM 576 MB/s bandwidth 4 memory ports (2 read, 1 write, 1 i/o) VME I/O subsystem 36 MFLOPS peak (100x100 is 9, 1000x1000 is 34) >Ditto for the YMP/EL. The YMP/EL has not, as of yet, been officially announced, but here are some specs which have been provided in the past: 30 ns clock 64-bit vector, Y-MP campatable 256 to 1024 MB memory 1 GB/sec bandwidth per CPU 1 to 4 CPUs 4 memory ports The rest will be available at time of announcement > >Thanks! > You're welcome! -- ******************************************************************************* * Dennis Ellis (dre@sequoia.cray.com) * * * Cray Research, Inc. * * * Eagan, MN 55121 * All the normal stuff about opinions.. *
uh311ae@sunmanager.lrz-muenchen.de (Henrik Klagges) (06/08/91)
Hello, just reading the specs of the small X-Cray (64 bit architecture, 36MFlops Peak, >.5GB/s mem bandwidth, 55 ns clock, 80 ns DRAM (64-128MB) ), I wonder how this machine compares to the new superworkstations like the snake from HP. These should get very close (60 MHz 64 bit architecture) at less than a tenth of the cost (e.g., 20-30k $). Cheers, Rick@vee.lrz-muenchen.de