[
https://issues.jboss.org/browse/ISPN-8550?page=com.atlassian.jira.plugin....
]
William Burns edited comment on ISPN-8550 at 11/27/17 3:41 PM:
---------------------------------------------------------------
I was using valgrind like the following:
{quote}
valgrind --tool=massif --time-unit=ms --detailed-freq=10000 <cmd>
{quote}
I increased detailed-freq really high to try to remove the excess.
I took the memory usage after it was primed with inserts and then subtracted that from the
peak memory usage. Looking at your spreadsheet I don't see excess JVM and others
subtracted from the size in your calculation. In my tests it was about 7M of memory used
by that, which can throw off your other numbers, or maybe I am not reading your
spreadsheet correctly.
Also I am not quite sure about the baseline value of 72M, seems quite high.
was (Author: william.burns):
I was using valgrind like the following:
{quote}
valgrind --tool=massif --time-unit=ms --detailed-freq=10000 <cmd>
{quote}
I increased detailed-freq really high to try to remove the excess.
I took the memory usage after it was primed with inserts and then subtracted that from the
peak memory usage. Looking at your spreadsheet I don't see excess JVM and others
subtracted from the size in your calculation. In my tests it was about 7M of memory used
by that, which can throw off your other numbers, or maybe I am not reading your
spreadsheet correctly.
Try to estimate malloc overhead and add to memory based eviction
----------------------------------------------------------------
Key: ISPN-8550
URL:
https://issues.jboss.org/browse/ISPN-8550
Project: Infinispan
Issue Type: Sub-task
Reporter: William Burns
Assignee: William Burns
Fix For: 9.2.0.Beta2, 9.1.4.Final
Attachments: allocs.ods
We should try to also estimate malloc overhead. We could do something like Dan mentioned
at
https://github.com/infinispan/infinispan/pull/5590#pullrequestreview-7805...
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)