AuroraMiddleware/skia2
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[perf] - Update docs.

Browse Source 
No-Try: true
Docs-Preview: https://skia.org/?cl=248811
Change-Id: I23deb4000dac155294f69429e37f675bcd3fa29c
Reviewed-on: https://skia-review.googlesource.com/c/skia/+/248811
Reviewed-by: Heather Miller <hcm@google.com>
Auto-Submit: Joe Gregorio <jcgregorio@google.com>
This commit is contained in:
Joe Gregorio 2019-10-17 14:22:27 -04:00
parent f272bea9a6
commit 5363719efb
4 changed files with 16 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

BIN
site/dev/testing/Cluster.png
View File

Binary file not shown.
Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 64 KiB

After

Width:  |  Height:  |  Size: 58 KiB

BIN
site/dev/testing/Perf.png
View File

Binary file not shown.
Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 68 KiB

After

Width:  |  Height:  |  Size: 169 KiB

BIN
site/dev/testing/Regression.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 64 KiB

33
site/dev/testing/skiaperf.md
View File

@ -1,40 +1,39 @@
Skia Perf
=========
[Skia Perf](https://perf.skia.org) is a Polymer-based web application for
analyzing and viewing performance metrics produced by Skia's testing
infrastructure.
[Skia Perf](https://perf.skia.org) is a web application for analyzing and
viewing performance metrics produced by Skia's testing infrastructure.
<img src=Perf.png style="margin-left:30px" align="left" width="800"/> <br clear="left">
Skia tests across a large number of platforms and configurations, and each
commit to Skia generates 240,000 individual values that are sent to Perf,
consisting mostly of performance benchmark results, but also including memory
and coverage data.
commit to Skia generates more than 400,000 individual values that are sent to
Perf, consisting mostly of performance benchmark results, but also including
memory and coverage data.
Perf includes tools for analyzing such a large corpus of data, the most
powerful is [k-means clustering](https://perf.skia.org/t/). This tool groups
large sets of performance metrics together based on how they change over time,
and highlights sets of metrics that have performance regressions.
Perf offers clustering, which is a tool to pick out trends and patterns in large sets of traces.
<img src=Cluster.png style="margin-left:30px" align="left" width="500"/> <br clear="left">
<img src=Cluster.png style="margin-left:30px" align="left" width="400"/> <br clear="left">
Calculations
------------
And can generate alerts when those trends spot a regression:
<img src=Regression.png style="margin-left:30px" align="left" width="800"/> <br clear="left">
## Calculations
Skia Perf has the ability to perform calculations over the test data
allowing you to build up interesting queries.
This query displays [the ratio of playback time in ms to the number of ops for desk\_wowwiki.skp](https://perf.skia.org/#1876):
This query displays the ratio of playback time in ms to the number of ops for desk\_wowwiki.skp:
ratio(
ave(fill(filter("name=desk_wowwiki.skp&sub_result=min_ms"))),
ave(fill(filter("name=desk_wowwiki.skp&sub_result=ops")))
)
You can also use the data to answer questions like [how many tests were run per commit](https://perf.skia.org/#1878).
You can also use the data to answer questions like how many tests were run per commit.
count(filter(""))
See Skia Perf for the [full list of functions available](https://perf.skia.org/help).
See Skia Perf for the [full list of functions available](https://perf.skia.org/help/).