[
https://issues.apache.org/jira/browse/HADOOP-13028?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15252474#comment-15252474
]
Steve Loughran commented on HADOOP-13028:
-----------------------------------------
I'd like counters in {{FileSystem}} of
Actions on individual blobs
# create
# stat
# copy
# delete
+ list (path)
In FS input stream:
* of times stream closed
* count of times aborted
* stream reopened due to forward seek, backwards seek;
* re-opened due to IO problem.
I'd also like these counters to be visible to tests; at the very least the
toString() operator should dump it, but ideally: raw counters. Why? Lets me
write tests which actually compare the no. of times actions take place (e.g.
forward-seek-closures()) and look at tuning the code for that, which can be
done a lot more deterministically than just measuring test duration in some
microbenchmark
> add counter and timer metrics for S3A HTTP & low-level operations
> -----------------------------------------------------------------
>
> Key: HADOOP-13028
> URL: https://issues.apache.org/jira/browse/HADOOP-13028
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: fs/s3, metrics
> Affects Versions: 2.8.0
> Reporter: Steve Loughran
> Priority: Minor
>
> against S3 (and other object stores), opening connections can be expensive,
> closing connections may be expensive (a sign of a regression).
> S3A FS and individual input streams should have counters of the # of
> open/close/failure+reconnect operations, timers of how long things take. This
> can be used downstream to measure efficiency of the code (how often
> connections are being made), connection reliability, etc.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
