[
https://issues.apache.org/jira/browse/HDFS-6982?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14125893#comment-14125893
]
Maysam Yabandeh commented on HDFS-6982:
---------------------------------------
bq. If this is the case, maybe we can separate the rolling window in a separate
jira. What do you think?
Sounds good to me. It would also splits the patch and make it easier to review.
So, to summarize, we are planning for three patches (in order):
# RollingWindow and RollingWindowManager
# TopMetrics and TopAuditLogger
# Minimal changes that hook nntop to the rest of nn
Each step is perquisite of the next.
This plan will still place the rolling window source code inside the nn, so I
am not sure whether your previous concern of the right place for rolling window
is addressed or not?
> 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, 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)
