Monday Jul 01, 2013

IBM "per core" comparisons for SPECjEnterprise2010

Responding to an IBM claim of double performance per core, questions and concerns are raised about:

  • Scaling is not free
  • Choosing the denominator radically changes the picture
  • SPEC Fair Use requirements
  • Substantiation and transparency
  • T5 claim for "fastest processor"
[Read More]

About this blog

Occasional discussion of computer performance, especially performance of CPU, Memory, and Compiler (the components measured by SPEC CPU)

I contributed to the development of SPEC CPU2006 and SPEC CPU2000, and serve as run rules editor for the CPU subcommittee.

In the picture, I am the one on the right. The file name for the picture says it all - I need to get out on that bike more often.

Old Fat Guy With One Foot In The Gravy

Search

Categories
Archives
« July 2013 »
SunMonTueWedThuFriSat
 
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
   
       
Today