Created a ticket: Improve HLC implementation performance (
https://issues.apache.org/jira/browse/HBASE-18686).
On Fri, Aug 25, 2017 at 11:53 AM, Enis Söztutar wrote:
> Indeed it seems like a perf improvement in the code HLC clock is needed
> before the merge. Do you mind
[
https://issues.apache.org/jira/browse/HBASE-18687?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack resolved HBASE-18687.
---
Resolution: Fixed
Pushed to branch-2 and master.
> Add @since 2.0.0 to new classes
>
Chia-Ping Tsai created HBASE-18690:
--
Summary: Replace o.a.h.c.InterfaceAudience by
o.a.h.h.c.InterfaceAudience
Key: HBASE-18690
URL: https://issues.apache.org/jira/browse/HBASE-18690
Project: HBase
It's true that hlc clock is very slower than system, but is it really a
bottleneck? I mean 7mil ops/sec with 16 threads doesn't sound bottleneck
because:
- It's like 70 times of any RS' throughput (assuming 100k write throughput)
- It's certainly enough for meta operations.
What say?
But i
huaxiang sun created HBASE-18693:
Summary: adding an option to restore_snapshot to move mob files
from archive dir to working dir
Key: HBASE-18693
URL: https://issues.apache.org/jira/browse/HBASE-18693
stack created HBASE-18691:
-
Summary: [compat 1-2] HCD remove and removeConfiguration change
return type
Key: HBASE-18691
URL: https://issues.apache.org/jira/browse/HBASE-18691
Project: HBase
Issue
stack created HBASE-18692:
-
Summary: [compat 1-2] ByteBufferUtils.copyFromBufferToBuffer goes
from void to int
Key: HBASE-18692
URL: https://issues.apache.org/jira/browse/HBASE-18692
Project: HBase
Amit Patel created HBASE-18686:
--
Summary: Improve HLC implementation performance
Key: HBASE-18686
URL: https://issues.apache.org/jira/browse/HBASE-18686
Project: HBase
Issue Type: Sub-task
Ted Yu created HBASE-18689:
--
Summary: BackupClientFactory class misses InterfaceAudience
annotation
Key: HBASE-18689
URL: https://issues.apache.org/jira/browse/HBASE-18689
Project: HBase
Issue
Appy created HBASE-18688:
Summary: Upgrade commons-codec to 1.10
Key: HBASE-18688
URL: https://issues.apache.org/jira/browse/HBASE-18688
Project: HBase
Issue Type: Task
Reporter: Appy
HLC will never be able to match the throughput for the system clock of
course. I was expecting the perf to be in the 2-3x slower range, not more
than an order of magnitude difference. The current perf is definitely
acceptable for meta updates, but if we are recommending HLC for user
tables, we
[
https://issues.apache.org/jira/browse/HBASE-18681?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Duo Zhang resolved HBASE-18681.
---
Resolution: Duplicate
Finally I found that it is possible to remove the LegacyScanQueryMatcher in
[
https://issues.apache.org/jira/browse/HBASE-16320?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Duo Zhang resolved HBASE-16320.
---
Resolution: Won't Fix
Resolve as all sub tasks have been cancelled.
> Revisit scan semantics and
[
https://issues.apache.org/jira/browse/HBASE-16322?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Duo Zhang resolved HBASE-16322.
---
Resolution: Won't Fix
As said previously, we do have some usages which use raw scan and filter
[
https://issues.apache.org/jira/browse/HBASE-16323?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Duo Zhang resolved HBASE-16323.
---
Resolution: Won't Fix
In HBASE-18169 we decided to disable the custom StoreScanner creation from CP
Peter Somogyi created HBASE-18683:
-
Summary: Upgrade hbase to commons-math 3
Key: HBASE-18683
URL: https://issues.apache.org/jira/browse/HBASE-18683
Project: HBase
Issue Type: Improvement
Hi,
I am looking at the replication code, and it has this line:
this.queueSizePerGroup = this.conf.getInt("hbase.regionserver.maxlogs", 32);
This drives the queue size accepting more WAL files. But we have a
different computation now, using the heap size instead, so often have
a maxlogs > 32 in
[
https://issues.apache.org/jira/browse/HBASE-14910?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Appy resolved HBASE-14910.
--
Resolution: Duplicate
Fix Version/s: HBASE-18640
> Create base hbase-mapred and hbase-io modules, move
stack created HBASE-18687:
-
Summary: Add @since 2.0.0 to new classes
Key: HBASE-18687
URL: https://issues.apache.org/jira/browse/HBASE-18687
Project: HBase
Issue Type: Sub-task
Reporter:
bq. ...but if we are recommending HLC for user tables, we have to also do
end-to-end testing to demonstrate that perf is not affected by HLC.
Agree. I think the only way of doing it right now is running ycsb
with/without HLC.
If it highlights an issue, it means there is one (there can't be false
Bo Cui created HBASE-18694:
--
Summary: Ha mode master switch causes region not online
Key: HBASE-18694
URL: https://issues.apache.org/jira/browse/HBASE-18694
Project: HBase
Issue Type: Bug
[
https://issues.apache.org/jira/browse/HBASE-18454?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Minwoo Kang resolved HBASE-18454.
-
Resolution: Information Provided
Actually, This is HDFS issue.
Please check HDFS-12204.
>
[
https://issues.apache.org/jira/browse/HBASE-15374?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Duo Zhang resolved HBASE-15374.
---
Resolution: Fixed
Assignee: Duo Zhang
Thanks for the reminder [~chia7712].
Resolve as all
[
https://issues.apache.org/jira/browse/HBASE-18454?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Minwoo Kang resolved HBASE-18454.
-
Resolution: Duplicate
> RegionServer Do not close file descriptor when using shortcircuit
>
[
https://issues.apache.org/jira/browse/HBASE-18454?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Minwoo Kang reopened HBASE-18454:
-
> RegionServer Do not close file descriptor when using shortcircuit
>
Agreed this part of code can be improved.
On Fri, Aug 25, 2017 at 3:57 AM, Lars George wrote:
> Hi,
>
> I am looking at the replication code, and it has this line:
>
> this.queueSizePerGroup = this.conf.getInt("hbase.regionserver.maxlogs",
> 32);
>
> This drives the queue
Just as an update with regard to performance: getting the time from or
updating the hybrid logical clock is quite expensive (see clocks
microbenchmark
stack created HBASE-18685:
-
Summary: Update refguide release candidate making section; stale
and bad instruction
Key: HBASE-18685
URL: https://issues.apache.org/jira/browse/HBASE-18685
Project: HBase
+1 (non-binding)
Signatures and checksums good.
LICENSE question... do we actually bundle jquery, asciidoctor, and the orca
logo here? I didn't see them anywhere.
CHANGES is mostly blank... If we don't use it, get rid of it in the next
version.
Verified an hbase compile against this version,
+1 (non-binding)
Checksum, signature good.
Built hbase pointing to staging repo.
I agree with Mike's LICENSE and CHANGES suggestions.
Peter
On Thu, Aug 24, 2017 at 12:46 AM, Stack wrote:
> This is a minor update to hbase-thirdparty, our little side project of
> relocated
stack created HBASE-18684:
-
Summary: [hbase-thirdparty] Remove CHANGES, update NOTICES
Key: HBASE-18684
URL: https://issues.apache.org/jira/browse/HBASE-18684
Project: HBase
Issue Type: Task
On Fri, Aug 25, 2017 at 10:06 AM, Mike Drob wrote:
> +1 (non-binding)
>
> Signatures and checksums good.
>
> LICENSE question... do we actually bundle jquery, asciidoctor, and the orca
> logo here? I didn't see them anywhere.
> CHANGES is mostly blank... If we don't use it, get
Indeed it seems like a perf improvement in the code HLC clock is needed
before the merge. Do you mind creating a sub ticket for this.
Enis
On Fri, Aug 25, 2017 at 10:54 AM, Amit Patel
wrote:
> Just as an update with regard to performance: getting the time from or
>
33 matches
Mail list logo