Article View: comp.arch
Article #2884Cray divide speed
From: John D. McCalpin
Date: Sun, 29 Oct 1989 22:40
Date: Sun, 29 Oct 1989 22:40
20 lines
1207 bytes
1207 bytes
Does anyone have the figures handy for the peak speed of the Cray X/MP, the Cray Y/MP, and the Cray 2 for vector divide operations? I know that Cray uses an approximate reciprocal scheme, but I need to know how many cycles per result are required for a plain vector divide, i.e. DO 100 I=1,N A(I) = B(I)/C(I) 100 CONTINUE On the Cyber 205 and ETA-10, the number is 6 cycles per element, and I am trying to figure out if that is a representative number for vector machines. I am using this information to modify the work estimates for various implicit and explicit finite-difference schemes used in fluid dynamics. -- John D. McCalpin - mccalpin@masig1.ocean.fsu.edu mccalpin@scri1.scri.fsu.edu mccalpin@delocn.udel.edu
Technical Details
Message-ID:
<MCCALPIN.89Oct29174054@masig3.masig3.ocean.fsu.edu>Path:
arc1-ca.pugleaf.net!utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uflorida!stat!stat.fsu.edu!mccalpinShow Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!uflorida!stat!stat.fsu.edu!mccalpin From: mccalpin@masig3.masig3.ocean.fsu.edu (John D. McCalpin) Newsgroups: comp.arch,comp.sys.super,comp.unix.cray Subject: Cray divide speed Message-ID: <MCCALPIN.89Oct29174054@masig3.masig3.ocean.fsu.edu> Date: 29 Oct 89 22:40:54 GMT Sender: news@stat.fsu.edu Distribution: usa Organization: Supercomputer Computations Research Institute Lines: 20