[ 
https://issues.apache.org/jira/browse/CASSANDRA-17240?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17619692#comment-17619692
 ] 

Branimir Lambov commented on CASSANDRA-17240:
---------------------------------------------

{quote}It has been recently agreed that new or modified tests should be run 500 
times before commit...
{quote}
The flakes in the memtable size test are now fixed. [~adelapena], you can rerun 
the above if you wish; I have already run both flaky ones, 
{{MemtableSizeOffheapObjectsTest}} and {{{}MemtableSizeOffheapBuffersTest{}}}, 
thousands of times.
{quote}It seems like there are things that we're now ignoring (Owner, 
MemtableAllocator, and initial comparator in AbstractAllocatorMemtable; empty 
clusterings; and the measureDeepOmitShared()) and would push flush frequency 
down a bit, but then the changes in Columns potentially pushing flush frequency 
up. Do we have an idea of where that nets out?
{quote}
The {{Unmetered}} annotations only really affect the calculation the memtable 
size test does. The flush frequency can increase due to the columns objects, 
possibly by up to 10% in key-value workloads with a small amount of data per 
column.

I will leave the memtable precision parts in a separate commit so that, if we 
are unhappy with the effect, they can be independently rolled back and held off 
until a major release.

> CEP-19: Trie memtable implementation
> ------------------------------------
>
>                 Key: CASSANDRA-17240
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-17240
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Local/Memtable
>            Reporter: Branimir Lambov
>            Assignee: Branimir Lambov
>            Priority: Normal
>         Attachments: SkipListMemtable-OSS.png, TrieMemtable-OSS.png, 
> density_SG.html.gz, density_test_with_sharding.html.gz, latency-1_1-95.png, 
> latency-9_1-95.png, throughput_SG.png, throughput_apache.png
>
>          Time Spent: 13h 20m
>  Remaining Estimate: 0h
>
> Trie-based memtable implementation as described in CEP-19, built on top of 
> CASSANDRA-17034 and CASSANDRA-6936.
> The implementation is available in this 
> [branch|https://github.com/blambov/cassandra/tree/CASSANDRA-17240].



--
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