[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15567741#comment-15567741
]
ramkrishna.s.vasudevan commented on HBASE-15561:
I will close this JIRA linking
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15564682#comment-15564682
]
ramkrishna.s.vasudevan commented on HBASE-15561:
it was only one client.
Client on the
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15564510#comment-15564510
]
stack commented on HBASE-15561:
---
Were we doing any i/o? Or all was cached?
Client was on same machine as
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15564497#comment-15564497
]
stack commented on HBASE-15561:
---
Was the server under load when only one client pushing load?
> See how
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15563893#comment-15563893
]
Vladimir Rodionov commented on HBASE-15561:
---
Yes.
We have a JIRA: HBASE-15513
> See how G1GC
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15563851#comment-15563851
]
Enis Soztutar commented on HBASE-15561:
---
So as a result of this, we will enable chunk pool by
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15554857#comment-15554857
]
ramkrishna.s.vasudevan commented on HBASE-15561:
Yes. All tests have been performed with
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15553616#comment-15553616
]
Vladimir Rodionov commented on HBASE-15561:
---
I think we have got enough pro-pool pro-MSLAB
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15550918#comment-15550918
]
ramkrishna.s.vasudevan commented on HBASE-15561:
bq.Sure, were these tests with chunk
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15550103#comment-15550103
]
Vladimir Rodionov commented on HBASE-15561:
---
{quote}
But the overall take away is that MSLAB ON
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15547932#comment-15547932
]
ramkrishna.s.vasudevan commented on HBASE-15561:
Oops. Sorry. I read the other way. The
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15547921#comment-15547921
]
stack commented on HBASE-15561:
---
Trunk does not have pooling? We want that, right?
> See how G1GC works
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15547894#comment-15547894
]
ramkrishna.s.vasudevan commented on HBASE-15561:
That is because the doc case we had done
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15547885#comment-15547885
]
stack commented on HBASE-15561:
---
In doc. you had more GCs when MSLAB was on?
> See how G1GC works with
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15547684#comment-15547684
]
ramkrishna.s.vasudevan commented on HBASE-15561:
Ping for feedback/reviews. This topic
[
https://issues.apache.org/jira/browse/HBASE-15561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15544437#comment-15544437
]
ramkrishna.s.vasudevan commented on HBASE-15561:
Just wanted to update this JIRa with our
16 matches
Mail list logo
