[ 
https://issues.apache.org/jira/browse/CASSANDRA-13530?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16259253#comment-16259253
 ] 

Jason Brown commented on CASSANDRA-13530:
-----------------------------------------

Update on testing: it seems like periodic mode doesn't test {{CommitLogTest}} 
very well in the current state of things. As the existing test only exercised 
{{batch}} mode, and given the scope of this ticket is to add {{group}} mode, I 
propose to ignore (not commit) the {{PeriodicCommitLogTest}} and save that for 
a followup ticket as it's slightly outside the scope of the current ticket. 
I'll commit {{BatchCommitLogTest}} and {{GroupCommitLogTest}} to preserve the 
existing test for batch and add the test for group. wdyt, @Ariel?

{{CommitLogStressTest}} began to timeout in circleci because each {{@Test}} 
method tested all configuration (compression, encryption, plain) and mode 
variants (periodic, batch, and group). I refactored it to be a 
{{@Parameterized}} test class, like {{CommitLogTest}}, and that didn't help. 
What did fix it was upping the {{test.long.timeout}} in the {{build.xml}} to 15 
minutes from 10, as the timeouts apply to the test class as a whole, not 
individual test methods. I can revert the refactor and just leave the build 
file change on commit, or we can keep both as the refactor tidied up a few 
things in that test class (purely subjective statement) - wdyt?


> GroupCommitLogService
> ---------------------
>
>                 Key: CASSANDRA-13530
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-13530
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: Yuji Ito
>            Assignee: Yuji Ito
>             Fix For: 2.2.x, 3.0.x, 3.11.x
>
>         Attachments: GuavaRequestThread.java, MicroRequestThread.java, 
> groupAndBatch.png, groupCommit22.patch, groupCommit30.patch, 
> groupCommit3x.patch, groupCommitLog_noSerial_result.xlsx, 
> groupCommitLog_result.xlsx
>
>
> I propose a new CommitLogService, GroupCommitLogService, to improve the 
> throughput when lots of requests are received.
> It improved the throughput by maximum 94%.
> I'd like to discuss about this CommitLogService.
> Currently, we can select either 2 CommitLog services; Periodic and Batch.
> In Periodic, we might lose some commit log which hasn't written to the disk.
> In Batch, we can write commit log to the disk every time. The size of commit 
> log to write is too small (< 4KB). When high concurrency, these writes are 
> gathered and persisted to the disk at once. But, when insufficient 
> concurrency, many small writes are issued and the performance decreases due 
> to the latency of the disk. Even if you use SSD, processes of many IO 
> commands decrease the performance.
> GroupCommitLogService writes some commitlog to the disk at once.
> The patch adds GroupCommitLogService (It is enabled by setting 
> `commitlog_sync` and `commitlog_sync_group_window_in_ms` in cassandra.yaml).
> The difference from Batch is just only waiting for the semaphore.
> By waiting for the semaphore, some writes for commit logs are executed at the 
> same time.
> In GroupCommitLogService, the latency becomes worse if the there is no 
> concurrency.
> I measured the performance with my microbench (MicroRequestThread.java) by 
> increasing the number of threads.The cluster has 3 nodes (Replication factor: 
> 3). Each nodes is AWS EC2 m4.large instance + 200IOPS io1 volume.
> The result is as below. The GroupCommitLogService with 10ms window improved 
> update with Paxos by 94% and improved select with Paxos by 76%.
> h6. SELECT / sec
> ||\# of threads||Batch 2ms||Group 10ms||
> |1|192|103|
> |2|163|212|
> |4|264|416|
> |8|454|800|
> |16|744|1311|
> |32|1151|1481|
> |64|1767|1844|
> |128|2949|3011|
> |256|4723|5000|
> h6. UPDATE / sec
> ||\# of threads||Batch 2ms||Group 10ms||
> |1|45|26|
> |2|39|51|
> |4|58|102|
> |8|102|198|
> |16|167|213|
> |32|289|295|
> |64|544|548|
> |128|1046|1058|
> |256|2020|2061|



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

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

Reply via email to