Continuous profiling for analysis of CPU and memory usage, down to the line number and throughout time. Saving infrastructure cost, improving performance, and increasing reliability.
The UI has seen many improvements in this release, including a redesign of the time range picker, a new feature to highlight similar stacks when hovering on a node, and the replacement of loading spinners with skeleton placeholders. Now, the UI also displays CPU cores per second for sample count profiles to make them comparable with different underlying profiling durations.
Full Changelog: https://github.com/parca-dev/parca/compare/v0.20.0...v0.21.0
This release features tons of performance improvements, this is the lowest CPU and memory usage Parca server yet!
Highlights:
Apache Arrow is now the one and only format to read and write data in the internal storage.
Full Changelog: https://github.com/parca-dev/parca/compare/v0.19.0...v0.20.0
v0.19.0 is the biggest and most exciting Parca release yet!
Aside from that, there are countless performance improvements and bug fixes.
Bigint
instead of string
on the UI by @manojVivek in https://github.com/parca-dev/parca/pull/2934
Full Changelog: https://github.com/parca-dev/parca/compare/v0.18.0...v0.19.0
This release comes packed with performance improvements, enhancements and so many fixes to the UI and CI!
Happy profiling!
docker pull ghcr.io/parca-dev/parca:v0.18.0
Join our Discord server; Follow us on Twitter; Read the documentation.
Full Changelog: https://github.com/parca-dev/parca/compare/v0.17.0...v0.18.0
Features
Enhancements
Fixes
Full Changelog: https://github.com/parca-dev/parca/compare/v0.16.0...v0.17.0
docker pull ghcr.io/parca-dev/parca:v0.17.0
Join our Discord server; Follow us on Twitter; Read the documentation.
Full Changelog: https://github.com/parca-dev/parca/compare/v0.16.1...v0.16.2
Full Changelog: https://github.com/parca-dev/parca/compare/v0.16.0...v0.16.1
This release is a major milestone for the Parca project. We have removed the Merge
button, as we are now comfortable with the performance of these queries being run by default when querying data. We now felt comfortable after months of work improving the performance of the storage.
Many many more awesome changes, below is a more exhaustive changelog.
Happy profiling!
""
matcher values to NULL. @metalmatze in https://github.com/parca-dev/parca/pull/2616
Full Changelog: https://github.com/parca-dev/parca/compare/v0.15.0...v0.16.0
This release has a breaking change in the debuginfo upload protocol, which is only compatible with Parca Agent v0.11.0 or higher (more details in https://github.com/parca-dev/parca/pull/2248).
Filter by function
functionality added to ProfileView by @manojVivek in https://github.com/parca-dev/parca/pull/2062 https://github.com/parca-dev/parca/pull/2156 https://github.com/parca-dev/parca/pull/2236
flamegraph_table
report by @manojVivek in https://github.com/parca-dev/parca/pull/2271
TopTable
performance and compare
option added to benchmark runner by @manojVivek in https://github.com/parca-dev/parca/pull/2225
Full Changelog: https://github.com/parca-dev/parca/compare/v0.14.0...v0.15.0
This release features a few small improvements and bug fixes but the three main topics are:
Full Changelog: https://github.com/parca-dev/parca/compare/v0.13.0...v0.14.0