[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Summary: Fix merge of MVCC and SequenceID performance regression in
branch-1.0 for Increments (was: Fix merge
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Enis Soztutar updated HBASE-15031:
--
Fix Version/s: (was: 1.0.4)
1.0.3
> Fix merge of MVCC and SequenceID
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 15031.v8.master.patch
Forward port that only goes the 'fast path'; makes sense only if all WAL
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Release Note:
Increments can be 10x slower (or more) when there is high concurrency since
HBase 1.0.0
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 15031.v8.branch-1.0.patch
Minor change. Start mvcc write before memstore update and finish after.
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Release Note:
Increments can be 10x slower (or more) when there is high concurrency since
HBase 1.0.0
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 15031.v7.branch-1.0.patch
Patch that incorporates [~eclark] feedback (uncovered bug in Scan where
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Resolution: Fixed
Hadoop Flags: Reviewed
Fix Version/s: 1.1.3
Status: Resolved (was:
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 15031.v6.branch-1.0.patch
Any chance of a bit of a review? This is for hbase-1.0. Will do
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Fix Version/s: 1.0.4
> Fix merge of MVCC and SequenceID performance regression in branch-1.0
>
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 15031.v6.branch-1.0.patch
Retry patch. Last run failed like this:
Printing hanging tests
Hanging
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 15031.v6.branch-1.0.patch
I had TestRegionIncrement at 10 threads and had bumped it to 100 for
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Release Note:
Increments can be 10x slower (or more) when there is high concurrency since
HBase 1.0.0
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Release Note:
Increments can be 10x slower (or more) when there is high concurrency since
HBase 1.0.0
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 15031.v5.branch-1.0.patch
Noticed that TestAtomicOperation does a contended Increment test across
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 15031.v2.branch-1.0.patch
Address @eclark comment.
I then also moved the increment tests that are
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 15031.v3.branch-1.0.patch
Fix failing tests (they found good stuff)
> Fix merge of MVCC and
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 14460.v0.branch-1.0.patch
This 1.0 patch will be very different to what goes upstream in master
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Status: Patch Available (was: Open)
> Fix merge of MVCC and SequenceID performance regression in branch-1.0
>
[
https://issues.apache.org/jira/browse/HBASE-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-15031:
--
Attachment: 15031.v4.branch-1.0.patch
Fix more test failings.
There is some overlap in TestAtomicOperation
20 matches
Mail list logo