pcp
pcp copied to clipboard
performancecopilot
Performance Co-Pilot
Supports translation of pcp metrics into opentelemetry format. Supports opentelemetry protocol http json. Updated relevent qa tests; added one new one (1977) to test pcp2opentelemetry http functionality. Added man page...
Extension to the pmlogger utility to allow remote PCP archive creation using a push mechanism, based on HTTP and a PMWEBAPI /logger API, instead of the traditional pmcd pull mechanism....
Bumps [docker/build-push-action](https://github.com/docker/build-push-action) from 6.17.0 to 6.18.0. Release notes Sourced from docker/build-push-action's releases. v6.18.0 Bump @docker/actions-toolkit from 0.61.0 to 0.62.1 in docker/build-push-action#1381 [!NOTE] Build summary is now supported with Docker Build...
![dependabot[bot] avatar](https://avatars.githubusercontent.com/in/29110?v=4 "dependabot[bot] avatar"){kind=avatar}
Updated pcp dstat tool to not rely on hard coded mappings between instance filtering command line options and the specific plugin they apply to. Updated QA test #1207 and the...
Addition of OpenTelemetry PMDA - domain num 88. Supports ingestion of OpenTelemetry data and conversion to PCP metrics. PMDA supports dynamic metric addition and removal congruent to source entries in...
There are opportunities for tighter integration when using PCP Archives with Grafana, especially when using the 'archive-analysis' container image . Several PCP utilities (pmdumplog, pmloglabel, perhaps others?) output the start/end...
Statistics relating to the performance overheads from running eBPF programs are available in modern kernels via the bpf(2) syscall: ``` * BPF_ENABLE_STATS * Description * Enable eBPF runtime statistics gathering....
Currently redis pmda looks for a $host:$port combination for pulling the redis metrics. But it is possible that redis is using unix socket and not port like following: `redis 29103...
Nathan says: We currently provide per-NUMA-node stats from meminfo and numastat but not vmstat It would be nice to have also the stats from the per-NUMA-node vmstat file available, perhaps...
It would be nice if `pcp numastat` could report the same details as `numastat -n` and `numastat -mz`. Not sure if it would be feasible to report the same as...