[ 
https://issues.apache.org/jira/browse/CASSANDRA-19947?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17913454#comment-17913454
 ] 

Dmitry Konstantinov commented on CASSANDRA-19947:
-------------------------------------------------

I have compared the trunk version vs the latest state of the branch using the 
same cassandra stress test as here: CASSANDRA-20165, with no constraints 
defines.
It covers only a very basic scenario of course, more like a sanity, but the 
results are good, the difference is within a error margin.
like:
baseline:
{code:java}
Results:
Op rate                   :  161,201 op/s  [WRITE: 161,201 op/s]
Partition rate            :  161,201 pk/s  [WRITE: 161,201 pk/s]
Row rate                  :  161,201 row/s [WRITE: 161,201 row/s]
Latency mean              :    0.6 ms [WRITE: 0.6 ms]
Latency median            :    0.5 ms [WRITE: 0.5 ms]
Latency 95th percentile   :    0.9 ms [WRITE: 0.9 ms]
Latency 99th percentile   :    1.3 ms [WRITE: 1.3 ms]
Latency 99.9th percentile :    9.9 ms [WRITE: 9.9 ms]
Latency max               :  135.4 ms [WRITE: 135.4 ms]
Total partitions          : 10,000,000 [WRITE: 10,000,000]
Total errors              :          0 [WRITE: 0]
Total GC count            : 24
Total GC memory           : 86.798 GiB
Total GC time             :    1.5 seconds
Avg GC time               :   62.2 ms
StdDev GC time            :   17.2 ms
Total operation time      : 00:01:02
{code}
with the changes:
{code:java}
Results:
Op rate                   :  159,616 op/s  [WRITE: 159,616 op/s]
Partition rate            :  159,616 pk/s  [WRITE: 159,616 pk/s]
Row rate                  :  159,616 row/s [WRITE: 159,616 row/s]
Latency mean              :    0.6 ms [WRITE: 0.6 ms]
Latency median            :    0.6 ms [WRITE: 0.6 ms]
Latency 95th percentile   :    0.9 ms [WRITE: 0.9 ms]
Latency 99th percentile   :    1.2 ms [WRITE: 1.2 ms]
Latency 99.9th percentile :    7.9 ms [WRITE: 7.9 ms]
Latency max               :   95.7 ms [WRITE: 95.7 ms]
Total partitions          : 10,000,000 [WRITE: 10,000,000]
Total errors              :          0 [WRITE: 0]
Total GC count            : 27
Total GC memory           : 88.244 GiB
Total GC time             :    1.5 seconds
Avg GC time               :   57.0 ms
StdDev GC time            :   21.7 ms
Total operation time      : 00:01:02
{code}

I do not see new suspicious places related to CEP-42 in async profiler CPU 
profile:  [^cpu_profile_cep42.html] 

> CEP-42: Constraints Framework
> -----------------------------
>
>                 Key: CASSANDRA-19947
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-19947
>             Project: Apache Cassandra
>          Issue Type: Improvement
>          Components: CQL/Syntax
>            Reporter: Bernardo Botella
>            Assignee: Bernardo Botella
>            Priority: Normal
>         Attachments: ci_summary.html, cpu_profile_cep42.html
>
>          Time Spent: 24h 40m
>  Remaining Estimate: 0h
>
> Add new Constraints framework as described in CEP-42. This initial Jira 
> ticket includes the core constraints length and numeric. Follow up tickets 
> will include the rest of the constraints described in the CEP.
>  
> [https://cwiki.apache.org/confluence/display/CASSANDRA/CEP-42%3A+Constraints+Framework]



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to