Slow math performance on i7-5960X vs i7-3820
Hi, I'm running my own C++ developed neural net simulation software.I have two machines with an i7-5960X and an i7-3820 at similar clock speeds. The same code (single threaded) runs much more slowly...
View ArticleUnderstanding OFFCORE_RESPONSE_0 and 1
I am new to hardware counter measurements on intel. I want to measure L3 misses to local and remote DRAM on Intel Ivy bridge Model 62(Intel(R) Xeon(R) CPU E7-4860):I need to understand the difference...
View ArticleTsc-deadline feature. Cpu family ?
Hello,I know that the tsc-deadline ( != tsc) feature has been added on recent intel cpu.My question is: on which families ?Thanks !
View ArticleRDTSCP & cache misses
Hi all,I would like to benchmark some parts of my application using RDTSCP counters. My only concern with RDTSCP is if it takes into consideration L1/L2/LLC misses. The plan is to benchmark how long...
View ArticleCache Allocation Technology - Current specs (allocation granularity, number...
Hi, I am interested in testing out the new Intel Cache Allocation Technology. However I have been unable to find the specifications of CAT functionality provided by the 6 processors. I would like to...
View ArticleNUMA options in BIOS
Hi , i run my multi-thread application on a XEON E5_2620 that has NUMA on/off option in the BIOS. i profile my application using pcm-numa of intel PCM. when i disable hyperthreading as well as the...
View ArticleCan't download PCM 2.8
When I try to download PCM 2.8, starting from this page:https://software.intel.com/en-us/articles/intel-performance-counter-monitorI have to log in, then accept an open source license, and then finally...
View Article[PCM] building OS X driver fails
Danilcha D commented on the Intel PCM article::Hello!Could you please help me?I was building the OS X driver and it failed: Check dependencies [BEROR]error: There is no SDK with the name or path...
View Article[PCM] Kernel panic on OS X 10.9
Hello,My computer restarts every time I try to launch a simple program: PCM *m = PCM::getInstance(); if (m->program(PCM::DEFAULT_EVENTS, NULL) != PCM::Success) { std::cerr << "Failed to start...
View ArticleA curious bandwidth result
Now I am testing Stream Benchmark on a Numa System including two Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz. The compiler is GCC. I use a OpenMP version with 12 threads. I know that Numa system allocate...
View Article