Intel's Open-Source Technology Center has done their inaugural release of NumaTOP.
Per its 01.org project page
, "NumaTOP is an observation tool for runtime memory locality characterization and analysis of processes and threads running on a NUMA system. It helps the user characterize the NUMA behavior of processes and threads and identify where the NUMA-related performance bottlenecks reside."
Aside from needing this new user-space utility, for those wishing to do this NUMA memory profiling also need to patch their kernel with a perf load latency patch. This perf load support is expected to be merged into a future Linux kernel release.
More details on the NumaTOP development and release from Intel can be found within this kernel mailing list announcement
. "Performance analysis engineers know that NUMA can seriously impact performance and that NUMA performance analysis can be challenging. We've realized that currently there isn't an easy-to-use tool that lets us easily observe whether NUMA-related issues exist and, if so, where the NUMA bottleneck(s) reside. It can be quite challenging, especially in complex server environments."