hbase-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From lars hofhansl <la...@apache.org>
Subject Re: [ANNOUNCE] The 1st hbase 0.94.9 release candidate is available for download
Date Sat, 06 Jul 2013 16:50:32 GMT
Are these times or ops/sec, or something else?




----- Original Message -----
From: Jean-Marc Spaggiari <jean-marc@spaggiari.org>
To: dev@hbase.apache.org; lars hofhansl <larsh@apache.org>
Cc: 
Sent: Friday, July 5, 2013 8:24 PM
Subject: Re: [ANNOUNCE] The 1st hbase 0.94.9 release candidate is available
 for download

I have redo the tests and it's now all very consistent between all the
recent versions

Min    Max    Average    Min80%    Max80%    Variance    Precision
hbase-0.94.5    914    1175    929    920    944    172    18.46%
hbase-0.94.6    913    1176    927    915    954    174    18.80%
hbase-0.94.7    913    1295    930    919    946    170    18.30%
hbase-0.94.8    914    947    935    914    945    166    17.78%
hbase-0.94.9    921    947    936    933    940    164    17.55%

I will try to make the tests last longer to get more precise results...

If we think a few nodes clusters is way beter for the performances
tracking, then let's work that way. But if we think PE can do the job,
it's even better ;) I just want to make sure we have accurate results
and that we performance test the right things...

I will re-run that, update the pdf and upload it again...

JM

2013/7/4 lars hofhansl <larsh@apache.org>:
> Thanks for doing all this work!
>
> I did find a pretty high variance between the runs, which may be at issue here.
>
> Maybe there are effects that only show on certain setups (disk drives, ram, jvm version,
etc).
>
> Maybe we can compare setups offline.
>
> -- Lars
>
>
>
> ----- Original Message -----
> From: Jean-Marc Spaggiari <jean-marc@spaggiari.org>
> To: dev@hbase.apache.org
> Cc:
> Sent: Thursday, July 4, 2013 9:25 AM
> Subject: Re: [ANNOUNCE] The 1st hbase 0.94.9 release candidate is available
>  for download
>
> Ok. I get similar performances between 0.94.7 and 0.94.9 for the randomRead
>
> 0.94.7 => 1124179.9ms
> 0.94.9 => 1110489.3ms
>
> I will re-run the tests one by one for all the 0.94.x versions and
> update my charts...
>
> 2013/7/3 Jean-Marc Spaggiari <jean-marc@spaggiari.org>:
>> Ok. Test pass if I run it alone:
>>
>> mvn clean test -P runSmallTests -Djava.net.preferIPv4Stack=true
>> -Dtest=org.apache.hadoop.hbase.coprocessor.TestRegionObserverScannerOpenHook
>>
>> -------------------------------------------------------
>>  T E S T S
>> -------------------------------------------------------
>> Running org.apache.hadoop.hbase.coprocessor.TestRegionObserverScannerOpenHook
>> Tests run: 3, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 11.642 sec
>>
>> Results :
>>
>> Tests run: 3, Failures: 0, Errors: 0, Skipped: 0
>>
>>
>>
>> And finally passed on another one:
>> Results :
>>
>> Tests run: 1359, Failures: 0, Errors: 0, Skipped: 13
>>
>> [INFO] ------------------------------------------------------------------------
>> [INFO] BUILD SUCCESS
>> [INFO] ------------------------------------------------------------------------
>> [INFO] Total time: 1:33:34.925s
>> [INFO] Finished at: Tue Jul 02 16:37:59 EDT 2013
>> [INFO] Final Memory: 44M/594M
>> [INFO] ------------------------------------------------------------------------
>>
>>
>> So it's all +1 for me.
>>
>> JM
>>
>> 2013/7/2 Elliott Clark <eclark@apache.org>:
>>> +1
>>>
>>> untared.
>>> Checked the sig.
>>> Ran Ycsb.
>>> Ran PE.
>>> upgraded from 94.8.
>>>
>>> On Tue, Jul 2, 2013 at 11:32 AM, lars hofhansl <larsh@apache.org> wrote:
>>>> passes fine here too
>>>>
>>>>
>>>>
>>>> ----- Original Message -----
>>>> From: Jean-Daniel Cryans <jdcryans@apache.org>
>>>> To: "dev@hbase.apache.org" <dev@hbase.apache.org>
>>>> Cc:
>>>> Sent: Tuesday, July 2, 2013 11:22 AM
>>>> Subject: Re: [ANNOUNCE] The 1st hbase 0.94.9 release candidate is available
>>>>  for download
>>>>
>>>> Yup:
>>>>
>>>> Running org.apache.hadoop.hbase.coprocessor.TestRegionObserverScannerOpenHook
>>>> Tests run: 3, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 22.756 sec
>>>>
>>>> Have you looked at the log to see on what error it fails?
>>>>
>>>> J-D
>>>>
>>>> On Tue, Jul 2, 2013 at 11:08 AM, Jean-Marc Spaggiari
>>>> <jean-marc@spaggiari.org> wrote:
>>>>> JD, does TestRegionObserverScannerOpenHook work for you?
>>>>>
>>>>> I will retry to run it again on 3 different computers to see.
>>>>>
>>>>> JM
>>>>>
>>>>> 2013/7/1 lars hofhansl <larsh@apache.org>:
>>>>>> Hah. I let Cody answer here. I didn't like the empty table either,
but he preferred it.
>>>>>>
>>>>>> -- Lars
>>>>>>
>>>>>>
>>>>>>
>>>>>> ________________________________
>>>>>>  From: Jean-Daniel Cryans <jdcryans@apache.org>
>>>>>> To: "dev@hbase.apache.org" <dev@hbase.apache.org>; lars hofhansl
<larsh@apache.org>
>>>>>> Sent: Monday, July 1, 2013 10:35 AM
>>>>>> Subject: Re: [ANNOUNCE] The 1st hbase 0.94.9 release candidate is
available for download
>>>>>>
>>>>>>
>>>>>> I ran some PE single node tests on both tars, looked at the web UIs
>>>>>> and logs. I'm +1 on this RC.
>>>>>>
>>>>>> My only nit is the way that HBASE-5083 leaves an empty table if there
>>>>>> are not backup masters is kinda ugly.
>>>>>>
>>>>>> J-D
>>>>>>
>>>>>> On Mon, Jun 24, 2013 at 3:51 PM, lars hofhansl <larsh@apache.org>
wrote:
>>>>>>> The 1st 0.94.9 RC is available for download at http://people.apache.org/~larsh/hbase-0.94.9-rc0/
>>>>>>> Signed with my code signing key: C7CFE328
>>>>>>>
>>>>>>> Like the previous point releases, 0.94.9 is a primarily a bug
fix release. This RC is comparatively small with 26 issues resolved against it:
>>>>>>>     [HBASE-8453] - TestImportExport failing again due to configuration
issues
>>>>>>>     [HBASE-8494] - TestRemoteAdmin#testClusterStatus should
not assume 'requests' does not change
>>>>>>>     [HBASE-8522] - Archived hfiles and old hlogs may be deleted
immediately by HFileCleaner, LogCleaner in HMaster
>>>>>>>     [HBASE-8555] - FilterList correctness may be affected by
random ordering of sub-filter(list)
>>>>>>>     [HBASE-8590] - [0.94] BlockingMetaScannerVisitor should
check for parent meta entry while waiting for split daughter
>>>>>>>     [HBASE-8639] - Poor performance of htable#getscanner in
multithreaded environment due to DNS.getDefaultHost() being called in ScannerCallable#prepare()
>>>>>>>     [HBASE-8640] - ServerName in master may not initialize
with the configured ipc address of hbase.master.ipc.address
>>>>>>>     [HBASE-8655] - Backport to 94 - HBASE-8346(Prefetching
.META. rows in case only when useCache is set to true)
>>>>>>>     [HBASE-8656] - Rpc call may not be notified in SecureClient
>>>>>>>     [HBASE-8671] - Per-region WAL breaks CP backwards compatibility
in 0.94 for non-enabled case
>>>>>>>     [HBASE-8684] - Table Coprocessor can't access external
HTable by default
>>>>>>>     [HBASE-8700] - IntegrationTestBigLinkedList can fail due
to random number collision
>>>>>>>     [HBASE-8724] - [0.94] ExportSnapshot should not use hbase.tmp.dir
as a staging dir on hdfs
>>>>>>>     [HBASE-8742] - HTableDescriptor Properties not preserved
when cloning
>>>>>>>     [HBASE-8743] - upgrade hadoop-23 version to 0.23.7
>>>>>>>     [HBASE-8749] - Potential race condition between FSUtils.renameAndSetModifyTime()
and HFile/LogCleaner
>>>>>>>     [HBASE-8762] - Performance/operational penalty when calling
HTable.get with a list of one Get
>>>>>>>     [HBASE-8783] - RSSnapshotManager.ZKProcedureMemberRpcs
may be initialized with the wrong server name
>>>>>>>     [HBASE-5083] - Backup HMaster should have http infoport
open with link to the active master
>>>>>>>     [HBASE-8609] - Make the CopyTable support startRow, stopRow
options
>>>>>>>     [HBASE-8636] - Backport KeyValue Codec to 0.94 (HBASE-7413)
>>>>>>>     [HBASE-8683] - Add major compaction support in CompactionTool
>>>>>>>     [HBASE-8692] - [AccessController] Restrict HTableDescriptor
enumeration
>>>>>>>     [HBASE-8702] - Make WALEditCodec pluggable
>>>>>>>     [HBASE-8504] - HTable.getRegionsInRange() should provide
a non-cached API
>>>>>>>     [HBASE-8603] - Backport HBASE-6921 to 0.94
>>>>>>>
>>>>>>> The full list of changes is also available here: https://issues.apache.org/jira/secure/ReleaseNote.jspa?projectId=12310753&version=12324431
>>>>>>>
>>>>>>> Please try out the RC, check out the doc, take it for a spin,
etc, and vote +1/-1 by July 2nd on whether we should release this as 0.94.9.
>>>>>>>
>>>>>>> The release testing spreadsheet is available here:
>>>>>>> https://docs.google.com/spreadsheet/ccc?key=0AvLqcVIqNtlTdGFHaktDR0FqLWdtQWZwdncyVzE3Z2c#gid=0
>>>>>>> (if you test this release, please add your test to this spreadsheet,
so that we can gauge the coverage)
>>>>>>>
>>>>>>> Thanks.
>>>>>>>
>>>>>>> -- Lars
>>>>>>>
>>>>>>> ps. I will likely be without access to the Internet for the next
five days
>>>>
>


Mime
View raw message