cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ariel Weisberg (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (CASSANDRA-6976) Determining replicas to query is very slow with large numbers of nodes or vnodes
Date Wed, 05 Nov 2014 23:17:34 GMT

     [ https://issues.apache.org/jira/browse/CASSANDRA-6976?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Ariel Weisberg updated CASSANDRA-6976:
--------------------------------------
    Attachment: GetRestrictedRanges.java
                jmh_output_murmur3.txt

With the random partitioner and the nodes and vnodes properly mapped to 256 tokens per endpoint.
{quote}{{
     [java] Benchmark                                        (nodes)  Mode  Samples      Score
     Error  Units
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkGlobal          1  avgt       15     16.656
±    0.821  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkGlobal         10  avgt       15    117.900
±    7.569  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkGlobal         50  avgt       15    584.864
±   39.682  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkGlobal        100  avgt       15   1246.623
±   90.073  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkGlobal        500  avgt       15  11324.498
±  915.911  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkGlobal       1000  avgt       15  26093.448
± 1704.904  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkGlobal       2000  avgt       15  61366.412
± 3471.103  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkRange           1  avgt       15     16.271
±    0.829  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkRange          10  avgt       15    108.126
±    5.609  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkRange          50  avgt       15    531.494
±   21.751  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkRange         100  avgt       15    986.310
±   53.092  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkRange         500  avgt       15  10127.756
±  388.324  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkRange        1000  avgt       15  24782.748
± 2220.330  us/op
     [java] o.a.c.t.m.GetRestrictedRanges.benchmarkRange        2000  avgt       15  54076.290
±  952.472  us/op
}}{quote}

> Determining replicas to query is very slow with large numbers of nodes or vnodes
> --------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-6976
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-6976
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>            Reporter: Benedict
>            Assignee: Ariel Weisberg
>              Labels: performance
>             Fix For: 2.1.2
>
>         Attachments: GetRestrictedRanges.java, jmh_output.txt, jmh_output_murmur3.txt,
make_jmh_work.patch
>
>
> As described in CASSANDRA-6906, this can be ~100ms for a relatively small cluster with
vnodes, which is longer than it will spend in transit on the network. This should be much
faster.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message