[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15686989#comment-15686989
]
Tomek Rękawek commented on OAK-4882:
It makes perfect sense. I'll take care of this tomorrow.
>
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15686912#comment-15686912
]
Vikas Saurabh commented on OAK-4882:
[~tomek.rekawek], since OAK-2761 was back ported to 1.2 as well.
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15637471#comment-15637471
]
Tomek Rękawek commented on OAK-4882:
Enabled the async mode by default in OAK-5071.
> Bottleneck in the
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15632034#comment-15632034
]
Tomek Rękawek commented on OAK-4882:
[~chetanm], I think we can try it, at least for trunk. [~tmueller]
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15631894#comment-15631894
]
Chetan Mehrotra commented on OAK-4882:
--
[~tomek.rekawek] New numbers look good. Should we now enable
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15629029#comment-15629029
]
Tomek Rękawek commented on OAK-4882:
Using stats from OAK-5038 it is possible to measure how many
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15621825#comment-15621825
]
Tomek Rękawek commented on OAK-4882:
Implemented a simple benchmark in OAK-5035. The way to reproduce
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15614788#comment-15614788
]
Chetan Mehrotra commented on OAK-4882:
--
[~tomek.rekawek] It would be helpful to invest in some
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15614762#comment-15614762
]
Tomek Rękawek commented on OAK-4882:
I committed the async queue simplified as in the discussion above.
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15608112#comment-15608112
]
Tomek Rękawek commented on OAK-4882:
[~mmarth]:
bq. But the way you describe it, it seems that the file
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15605437#comment-15605437
]
Michael Marth commented on OAK-4882:
I see - thanks for explaining. In that case dropping put() seems
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15605371#comment-15605371
]
Tomek Rękawek commented on OAK-4882:
[~mmarth], the problem here is as follows:
a) in case of a high
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15605385#comment-15605385
]
Chetan Mehrotra commented on OAK-4882:
--
bq. remove the oldest or drop the most recent put() entries -
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15605142#comment-15605142
]
Michael Marth commented on OAK-4882:
I am not sure if dropping update/invalidate operations occasionally
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15605044#comment-15605044
]
Tomek Rękawek commented on OAK-4882:
[~tmueller], [~mmarth], great, thanks for the feedback.
So, I was
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15604861#comment-15604861
]
Michael Marth commented on OAK-4882:
fwiw: IMHO the persistent cache should be fully tuned to take
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15604808#comment-15604808
]
Thomas Mueller commented on OAK-4882:
-
Yes, I think in theory we don't need to support updating entries
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15604732#comment-15604732
]
Tomek Rękawek commented on OAK-4882:
I'm still working on reproducing the original issue in a controlled
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15594824#comment-15594824
]
Tomek Rękawek commented on OAK-4882:
[~chetanm], absolutely, we need more systematic way to reproduce
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15594800#comment-15594800
]
Chetan Mehrotra commented on OAK-4882:
--
[~tomek.rekawek] It would also be helpful to have a oak-run
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15594791#comment-15594791
]
Tomek Rękawek commented on OAK-4882:
The first possible improvement is to remove the {{synchronized}}
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15544731#comment-15544731
]
Tomek Rękawek commented on OAK-4882:
Thread dump, redacted by Chetan:
A read blocks on cache load
[
https://issues.apache.org/jira/browse/OAK-4882?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15544723#comment-15544723
]
Tomek Rękawek commented on OAK-4882:
//cc: [~chetanm], [~tmueller]
> Bottleneck in the asynchronous
23 matches
Mail list logo
