[
https://issues.apache.org/jira/browse/HTRACE-69?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14679429#comment-14679429
]
Wangda Tan commented on HTRACE-69:
----------------------------------
Hi [~cmccabe], [~iwasakims],
First thanks for doing all the great HTrace stuffs, I'm currently looking at
opportunities to add HTrace support into YARN/MR.
Contribute some use cases from our side: I found maybe it's important to
support a parent determine if child need trace or not. From YARN's perspective,
most important things are:
- Why my AM takes so much time to be allocated/launched
- Why containers take so much time to be allocated/launched
The above two items can show 95% of YARN program slowness issue. Definitely we
still need more information such as HDFS/HBase trace information for the other
5% cases, we can turn it on when we need it. This can also reduce overhead of
store traces. I prefer to always enable YARN's application sampler, so user can
query it easily for any jobs ran on YARN. And we can turn on more fine graind
traces on demand.
Basically, I think YARN is a little different from HDFS, for a single MR/YARN
job, there can be millions+ operations to HDFS, but a single HDFS operation
couldn't be that complex.
Thoughts?
Regards,
Wangda
> Filtering child spans by sampler
> --------------------------------
>
> Key: HTRACE-69
> URL: https://issues.apache.org/jira/browse/HTRACE-69
> Project: HTrace
> Issue Type: New Feature
> Reporter: Masatake Iwasaki
> Assignee: Masatake Iwasaki
>
> Trace#startSpan respect the sampler given as argument only when there is no
> ongoing span (i.e. when creating new root span).
> {code}
> public static TraceScope startSpan(String description, Sampler<TraceInfo>
> s, TraceInfo tinfo) {
> Span span = null;
> if (isTracing() || s.next(tinfo)) {
> {code}
> Adding API starting span if {{(isTracing() && s.next(tinfo))}} enables
> filtering of child spans.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
