[
https://issues.apache.org/jira/browse/MATH-637?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Phil Steitz updated MATH-637:
-----------------------------
Attachment: PerfUtils.patch
One thing to think about is whether we really want / need to store all of the
timing results in an array. The times array could get quite large if the
number of iterations is very large.
The attached patch prepares for eventual move to just streaming the execution
times by changing the statistical aggregator from DescriptiveStatistics to
SummaryStatistics. The latter does not store velues in memory. The patch also
replaces the array of DescriptiveStatistics describing method results with an
array of StatisticalSummary instances. The latter are immutable value objects
that just maintain values of statistics. These changes should probably be made
even if we decide to hold the full time array in storage, since the current
impl is needlessly storing the raw data again inside the DescriptiveStatistics
instances and all you really need to store is the result of the statistical
analysis.
> Simple utility for micro-benchmarking
> -------------------------------------
>
> Key: MATH-637
> URL: https://issues.apache.org/jira/browse/MATH-637
> Project: Commons Math
> Issue Type: Wish
> Reporter: Gilles
> Assignee: Gilles
> Priority: Minor
> Labels: test
> Fix For: 3.0
>
> Attachments: PerfUtils.patch
>
>
> As per the discussion on the "dev" ML (with subject: "Simple benchmarking
> utility"), it would be useful to have an easy way to benchmark alternative
> implementations of some functionality.
> The proposed utility methods will be collected in a class located in the
> "src/test/java" part of the code repository.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
