[
https://issues.apache.org/jira/browse/TS-314?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12864102#action_12864102
]
John Plevyak commented on TS-314:
---------------------------------
I would hazard a guess that this was because of the very small partitions and
the way that objects were mapped to a partition and therefore ended up in a
single smallish RAM cache. I would imagine that if this was tested today with
a large RAM cache and a small number of partitions that the problem would not
exist.
> RAM cache only caches the first part of the file
> ------------------------------------------------
>
> Key: TS-314
> URL: https://issues.apache.org/jira/browse/TS-314
> Project: Traffic Server
> Issue Type: Improvement
> Reporter: Miles Libbey
> Priority: Minor
>
> (was yahoo bug 1562050)
> Original description
> by Bryan Call 2 years ago at 2007-10-26 08:53
> Leif was seeing a problem for large content, it was not all being served from
> RAM cache. Only the first 128K or
> something like that was being served. We should verify that RAM cache is
> working even for the maximum size objects.
>
>
> Comment 1
> by Leif Hedstrom 2 years ago at 2007-10-26 09:29:10
> I think it was more like ~32MB served out of RAM cache. I don't have a good
> number, the only way I noticed this was by
> looking at disk I/O while fetching one request from cache (and I can see disk
> I/O kicks in after somewhere around 32MB
> of fetched data).
>
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
