[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15384229#comment-15384229
]
Anoop Sam John commented on HBASE-14613:
Ya with CMS, the pooling was any way helpful only. With
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15383601#comment-15383601
]
ramkrishna.s.vasudevan commented on HBASE-14613:
bq.What about the CMS vs G1GC discussion
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15382553#comment-15382553
]
stack commented on HBASE-14613:
---
What about the CMS vs G1GC discussion above? You think it no longer
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15382387#comment-15382387
]
Anoop Sam John commented on HBASE-14613:
As part of the off heaping effort, we have added support
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15381960#comment-15381960
]
Lars Hofhansl commented on HBASE-14613:
---
What do we do with this? Either we do something or I'll
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15219386#comment-15219386
]
Jingcheng Du commented on HBASE-14613:
--
Supplement the G1GC settings used when I did the test in
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15214984#comment-15214984
]
Elliott Clark commented on HBASE-14613:
---
bq.No, no, not set to lower, instead, calculate the size
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15212845#comment-15212845
]
Lars Hofhansl commented on HBASE-14613:
---
Didn't think my musing about G1 and MSLAB would lead to a
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15212846#comment-15212846
]
Lars Hofhansl commented on HBASE-14613:
---
[~apurtell], [~vik.karma], maybe we should redo our G1
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15212814#comment-15212814
]
Yanping Wang commented on HBASE-14613:
--
Hi, Elliott
>>-XX:InitiatingHeapOccupancyPercent
>>Setting
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15212795#comment-15212795
]
ramkrishna.s.vasudevan commented on HBASE-14613:
Even mine was a default setting. I will
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15212078#comment-15212078
]
stack commented on HBASE-14613:
---
Excellent.
Woud you mind making a JIRA with a patch with your suggested
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15212076#comment-15212076
]
Elliott Clark commented on HBASE-14613:
---
bq.Jingcheng, pointed this JIRA to me. I am kind of
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15212023#comment-15212023
]
Vladimir Rodionov commented on HBASE-14613:
---
That was my first guess too - G1GC with default
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15211482#comment-15211482
]
Yanping Wang commented on HBASE-14613:
--
Yes, I worked on Jave runtime and GC performance for many
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15211464#comment-15211464
]
Yanping Wang commented on HBASE-14613:
--
Hi, Ram, The max size for a G1 region is 32MB, if you don't
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15211463#comment-15211463
]
stack commented on HBASE-14613:
---
[~ywang261] Thank you for jumping in. You figured the back-to-back marking
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15211455#comment-15211455
]
Yanping Wang commented on HBASE-14613:
--
Jingcheng, pointed this JIRA to me. I am kind of curious,
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15209735#comment-15209735
]
Vladimir Rodionov commented on HBASE-14613:
---
I am curios have anybody tried to run HBase with
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15209719#comment-15209719
]
ramkrishna.s.vasudevan commented on HBASE-14613:
Thanks [~eclark]. I think it would be
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15208690#comment-15208690
]
Elliott Clark commented on HBASE-14613:
---
bq.What could be the reason for that?
We did a lot of
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15208316#comment-15208316
]
ramkrishna.s.vasudevan commented on HBASE-14613:
bqThat means that if you're promoting a
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15208248#comment-15208248
]
ramkrishna.s.vasudevan commented on HBASE-14613:
I conducted a simple set of tests with
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207958#comment-15207958
]
Jingcheng Du commented on HBASE-14613:
--
Supplement some more information. The gc is measured as per
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207955#comment-15207955
]
Jingcheng Du commented on HBASE-14613:
--
I did some similar tests (YCSB, 1TB dataset, 1 HMaster and 3
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207933#comment-15207933
]
Anoop Sam John commented on HBASE-14613:
When working with HBASE-15180, discussed abt the check
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207901#comment-15207901
]
Lars Hofhansl commented on HBASE-14613:
---
[~vik.karma], FYI. Interesting bit here about disabling
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207781#comment-15207781
]
Anoop Sam John commented on HBASE-14613:
Thanks for sharing the details on what you see with
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207757#comment-15207757
]
Lars Hofhansl commented on HBASE-14613:
---
Not sure what you are referring to. SLAB making more
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207463#comment-15207463
]
Elliott Clark commented on HBASE-14613:
---
We've actually found that everything that tries to keep
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207460#comment-15207460
]
Elliott Clark commented on HBASE-14613:
---
Yeah g1 no mslab.
G1 really really loves having a balance
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207441#comment-15207441
]
stack commented on HBASE-14613:
---
You run G1 and no MSLAB?
What is the block cache issue? Is it L1 or L2 or
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207401#comment-15207401
]
Elliott Clark commented on HBASE-14613:
---
bq.Lately I've also been debating trying G1 without any
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15207369#comment-15207369
]
Vladimir Rodionov commented on HBASE-14613:
---
In other words, let us do more garbage, GC will
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15091531#comment-15091531
]
Lars Hofhansl commented on HBASE-14613:
---
Should I close this one?
Lately I've also been debating
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14972390#comment-14972390
]
Lars Hofhansl commented on HBASE-14613:
---
Hmm... Nothing in this thread represented a convincing
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14972393#comment-14972393
]
stack commented on HBASE-14613:
---
Smile.
bq. Nothing in this thread represented a convincing argument that
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14970461#comment-14970461
]
stack commented on HBASE-14613:
---
Yeah, I think we should keep memstorechunkpool -- not reallocating is
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14968181#comment-14968181
]
Enis Soztutar commented on HBASE-14613:
---
Nice analysis.
Would it help to simplify the code by
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14961041#comment-14961041
]
stack commented on HBASE-14613:
---
Now I don't trust what this program is showing us. I wanted to find where
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14961462#comment-14961462
]
Andrew Purtell commented on HBASE-14613:
bq. I am always dubious about trying to be smarter than
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14958401#comment-14958401
]
Lars Hofhansl commented on HBASE-14613:
---
More data is always good. :)
I doubt it'll make a
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14958387#comment-14958387
]
stack commented on HBASE-14613:
---
I was going to give it a run on the rig w/ it enabled to see if it
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14959810#comment-14959810
]
stack commented on HBASE-14613:
---
Nice link [~enis]
Looks like 8M in CMS ... or somewhere between 4M and
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14960206#comment-14960206
]
Lars Hofhansl commented on HBASE-14613:
---
I am always dubious about trying to be smarter than the
[
https://issues.apache.org/jira/browse/HBASE-14613?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14959381#comment-14959381
]
Enis Soztutar commented on HBASE-14613:
---
Echoing my comments from the other thread:
Do we know
46 matches
Mail list logo
