Benchmark Dalvik, Hard-Float OpenJDK and JAVA SE Startup Time
Why?
Benchmarking startup time of SciMark with the hard-float OpenJDK for ARM, Dalvik and JAVA SE on the same platform is a means for approximating the amount of headroom available to optimize in Dalivk startup time.
Context?
This is part of the larger ARM benchmarking effort and to inform a possible Dalvik optimization effort.
What gets produced?
A set of measurements on a Wiki, documentation for producing those measurements and a script to automate the measurements.
Where will the work get put?
Th scripts will be put in git and the docs on the Wiki or Google Doc.
Blueprint information
- Status:
- Not started
- Approver:
- Zach Pfeffer
- Priority:
- High
- Drafter:
- vishal
- Direction:
- Approved
- Assignee:
- vishal
- Definition:
- New
- Series goal:
- Accepted for future
- Implementation:
-
Unknown
- Milestone target:
-
backlog
- Started by
- Completed by
Related branches
Related bugs
Sprints
Whiteboard
Notes:
[2012/10/21 pfefferz] Please put notes here.
[2012/11/14 vishalbhoj] From the discussions at connect it came up that there could be design level changes in Dalvik hence we should not enter into the big ticket work of dalvik changes. Instead target architectural specific improvements. Hence moving this BP to backlog.
[2012/11/14 asac]: do we have a roadmap card for these efforts?
[2012/11/19 pfefferz] There'snot card for this at the moment. It was done to support our Connect discussions at the request of Michael Hope. We'll put it in the backlog fr now.
Meta:
Headline: SciMark startup time has been benchmarked against Dalvik, Hard-Float OpenJDK and JAVA SE.
Acceptance:
1. SciMark startup time has been benchmarked against Dalvik
2. SciMark startup time has been benchmarked against Hard Float OpenJDK
3. SciMark startup time has been benchmarked against JAVA SE
4. Scripts have been written that run the benchmarks
5. Results have been published
Work Items
Work items:
Figure out how to profile startup time: TODO
Document Dalivik, OpenJDK and JAVA SE starup: TODO
Write scripts to automate the measurements: TODO
Publish the results: TODO
Blog and promote results: TODO
Integrate results into Connect planning: TODO
