[ 
https://issues.apache.org/jira/browse/CASSANDRA-9067?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16326721#comment-16326721
 ] 

Jay Zhuang commented on CASSANDRA-9067:
---------------------------------------

Here is the patch, please review:
| Branch | uTest |
| [9067|https://github.com/cooldoger/cassandra/tree/9067] | 
[!https://circleci.com/gh/cooldoger/cassandra/tree/9067.svg?style=svg!|https://circleci.com/gh/cooldoger/cassandra/tree/9067]
 |

Here is the microbench test result (Score number is the running time, smaller 
is better):
{noformat}
     [java] Benchmark                                     (numElemsInK)  
(oldBfFormat)  Mode  Cnt      Score       Error  Units
     [java] BloomFilterSerializerBench.serializationTest              1         
  true  avgt    4    143.170 ?    35.566  us/op
     [java] BloomFilterSerializerBench.serializationTest              1         
 false  avgt    4     95.692 ?    61.473  us/op
     [java] BloomFilterSerializerBench.serializationTest             10         
  true  avgt    4    662.022 ?   202.386  us/op
     [java] BloomFilterSerializerBench.serializationTest             10         
 false  avgt    4    186.326 ?    87.105  us/op
     [java] BloomFilterSerializerBench.serializationTest            100         
  true  avgt    4   6312.909 ?  2090.136  us/op
     [java] BloomFilterSerializerBench.serializationTest            100         
 false  avgt    4   1125.158 ?   553.381  us/op
     [java] BloomFilterSerializerBench.serializationTest           1024         
  true  avgt    4  60436.370 ? 16174.150  us/op
     [java] BloomFilterSerializerBench.serializationTest           1024         
 false  avgt    4  10781.518 ?  6975.222  us/op
{noformat}

> BloomFilter serialization format should not change byte ordering
> ----------------------------------------------------------------
>
>                 Key: CASSANDRA-9067
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-9067
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: Benedict
>            Assignee: Jay Zhuang
>            Priority: Minor
>             Fix For: 4.x
>
>
> As a follow-up to CASSANDRA-9066 and CASSANDRA-9060, it appears we do some 
> unnecessary byte swapping during the serialization of bloom filters, which 
> makes the logic slower and harder to follow. We should either perform them 
> more efficiently (using Long.reverseBytes) or, preferably, eliminate the 
> conversion altogether since it does not appear to serve any purpose.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org

Reply via email to