Nehalem performance [Was: Is R more heavy on memory or processor?]
On Aug 27, 2009, at 11:16 , John C. Tull wrote:
On Aug 27, 2009, at 7:36 AM, Simon Urbanek wrote:
The tests used are from http://r.research.att.com/benchmarks (R 2.9.2 release was used but admittedly the OS version varied) Clearly, benchmarks never tell the full story and there may be uses that take advantage of the one or another architecture, but the bottom line is that Nehalems are not universally faster, so don't throuw your Harpertowns out just yet ;).
What was the memory configuration for these test systems? I would expect any differences there could account for significant variation as well.
Yes, you're right, but they are roughly comparable (Nehalem 12GB, Harpertown 2.8GHZ 10GB, Harpertown 2.66GHz 16GB). The test should not be reaching their limits - I was also running the 2.66 Harpertown with only 1GB of RAM with no big differences ... Cheers, Simon