- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello!
I'm trying to get the RAM and cache bandwidth metrics. It is seem possible to benchmark the system capacity, using lmbench for example, but I wonder how to do it for existing process (i.e. utilization)?
I see, that I can use uncore counters with `perf` utility, but seems like there is no uncore support on my processor:
model : 60
model name : Intel(R) Xeon(R) CPU E3-1230 v3 @ 3.30GHz
stepping : 3
microcode : 0x1a
# ls /sys/bus/event_source/devices/
breakpoint cpu software tracepoint
How can I track for the memory bandwidth metrics for some PID, more or less accurately?
Thanks in advance!
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
There are some memory bandwidth counters in the Xeon E3-1230 v3, but they are not easy to use....
Documentation is at https://software.intel.com/en-us/articles/monitoring-integrated-memory-controller-requests-in-the-2nd-3rd-and-4th-generation-intel
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Intel Performance Counter Monitor contains a sample implementation for reading these counters. However, they don't allow you to distinguish between processes. My recommendation is therefore that you do the measurement on an otherwise idle system.

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page