[
https://issues.apache.org/jira/browse/HADOOP-6623?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Todd Lipcon updated HADOOP-6623:
--------------------------------
Attachment: hadoop-6623.txt
Attached is an updated patch which adds a main() function to TestStringUtils
which acts as a benchmark. Benchmark results;
Java impl #4:1339ms
Java impl #5:1095ms
Java impl #6:1257ms
Java impl #7:1386ms
Java impl #8:1470ms
Java impl #9:1467ms
StringUtils impl #4:274ms
StringUtils impl #5:274ms
StringUtils impl #6:274ms
StringUtils impl #7:277ms
StringUtils impl #8:289ms
StringUtils impl #9:291ms
If I double the number of separators in the test string to 10, results are:
Java impl #4:1407ms
Java impl #5:1411ms
Java impl #6:1449ms
Java impl #7:1443ms
Java impl #8:1641ms
Java impl #9:1409ms
StringUtils impl #4:347ms
StringUtils impl #5:347ms
StringUtils impl #6:346ms
StringUtils impl #7:347ms
StringUtils impl #8:355ms
StringUtils impl #9:346ms
> Add StringUtils.split for non-escaped single-character separator
> ----------------------------------------------------------------
>
> Key: HADOOP-6623
> URL: https://issues.apache.org/jira/browse/HADOOP-6623
> Project: Hadoop Common
> Issue Type: Improvement
> Components: util
> Affects Versions: 0.22.0
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
> Priority: Minor
> Attachments: hadoop-6623.txt, hadoop-6623.txt
>
>
> This is for HDFS-1028 but useful generally. String.split("/") for example is
> way slower than an implementation that is specific to only single-character
> separators.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
