[
https://issues.apache.org/jira/browse/HDFS-6982?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14209024#comment-14209024
]
Andrew Wang commented on HDFS-6982:
-----------------------------------
How would this behave to a sudden, large spike in operations? This is the
situation we're trying to detect. i.e. for something like:
{noformat}
0, 0, 0, 1000000, 0, 0, 0, ...
{noformat}
What I'd want to see is essentially a step function going 0 -> 1000000 -> 0,
but an EWMA would necessarily tail off exponentially.
I'm also happy to take a look at any references you have. I've done some
reading on calculating percentiles on rolling windows, and what we have now is
pretty typical for that, i.e. a number of buckets each representing a fixed
time interval, aggregating buckets to calculate the metric, old buckets being
discarded as time passes.
> nntop: top-like tool for name node users
> -----------------------------------------
>
> Key: HDFS-6982
> URL: https://issues.apache.org/jira/browse/HDFS-6982
> Project: Hadoop HDFS
> Issue Type: New Feature
> Reporter: Maysam Yabandeh
> Assignee: Maysam Yabandeh
> Attachments: HDFS-6982.patch, HDFS-6982.v2.patch, HDFS-6982.v3.patch,
> HDFS-6982.v4.patch, HDFS-6982.v5.patch, HDFS-6982.v6.patch,
> nntop-design-v1.pdf
>
>
> In this jira we motivate the need for nntop, a tool that, similarly to what
> top does in Linux, gives the list of top users of the HDFS name node and
> gives insight about which users are sending majority of each traffic type to
> the name node. This information turns out to be the most critical when the
> name node is under pressure and the HDFS admin needs to know which user is
> hammering the name node and with what kind of requests. Here we present the
> design of nntop which has been in production at Twitter in the past 10
> months. nntop proved to have low cpu overhead (< 2% in a cluster of 4K
> nodes), low memory footprint (less than a few MB), and quite efficient for
> the write path (only two hash lookup for updating a metric).
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
