Performance Co-Pilot is a system performance analysis toolkit.
Analyze systems' performance metrics in real-time or using historical data.
Compare performance metrics between different hosts and different intervals. Observe trends and identify abnormal patterns.
Extend the collected performance metrics in a simple way.
PCP offers a multitude of APIs and libraries to extract and make use of performance metrics from your own application.
Sep 2018
PCP 4.1.3
Support for PostgreSQL 10.x in a new python pmdapostgresql. Labels support in pmlogrewrite and improvements to other utilities like pcp-dstat. Improvements to the BCC, KVM, Redis and Linux kernel PMDAs.
Aug 2018
PCP 4.1.1
Jun 2018
PCP 4.1.0
A colourful new pcp-dstat utility, building on the dstat legacy. Archive discovery and log tailing APIs. eBPF/BCC PMDA additions - new modules, histograms. Kernel metrics - AIO, interrupts, socket statistics, file locking, OOM, session count metrics. Automatic metadata compression.
May 2018
PCP 4.0.2
Further archive management improvements. New pcp-pidstat process tracking features. Performance improvements and label filtering in the Prometheus PMDA. The nfsclient PMDA rewritten in python. New SMART PMDA for disk health monitoring. Better handling of compressed archives in pmwebd. Numerous important SELinux updates and bug fixes in libpcp.
March 2018
PCP 4.0.1
Archive management improvements allowing fine-grained compression. New shell-based PCP service discovery. New daily reporting utility, pmlogger_daily_report. Scalability improvements and metric filtering in the Prometheus PMDA. The PostgreSQL PMDA now supports up to version 9.6.