[
https://issues.apache.org/jira/browse/CASSANDRA-19947?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17913436#comment-17913436
]
Bernardo Botella Corbi commented on CASSANDRA-19947:
----------------------------------------------------
Here is another interesting experiment. Things start to break when the test
gets to 250K req/second. The good news is that both trunk and CEP-42 with a
constraint codebases break more or less the same (error rates vary between
runs. For instance, in my last run below trunk behaved worse than CEP-42 code).
Having said that, both p99 latencies remain pretty similar (yay!) and error
rates are not very different either (yay!)
{code:java}
Trunk
Writes Reads
Deletes Errors
Count Latency (p99) 1min (req/s) | Count Latency (p99) 1min (req/s)
| Count Latency (p99) 1min (req/s) | Count 1min (errors/s)
162250910 51.42 240639.99 | 0 0 0
| 0 0 0 | 3668872 6834.57
162894818 51.42 241396.92 | 0 0 0
| 0 0 0 | 3668872 6398.83
163747828 51.12 241396.92 | 0 0 0
| 0 0 0 | 3718705 6398.83
164532038 62.83 242846.79 | 0 0 0
| 0 0 0 | 3770084 7036.96
165229862 61.78 242986.14 | 0 0 0
| 0 0 0 | 3795444 7348.6
165902297 64.35 242986.14 | 0 0 0
| 0 0 0 | 3804374 7348.6
166670259 65.77 243338.83 | 0 0 0
| 0 0 0 | 3827988 7281.45
167436232 61.76 243338.83 | 0 0 0
| 0 0 0 | 3860766 7281.45
168139952 61.31 243431.92 | 0 0 0
| 0 0 0 | 3889349 7223.41
168856033 61.76 242625.53 | 0 0 0
| 0 0 0 | 3892249 7149.31
Writes Reads
Deletes Errors
Count Latency (p99) 1min (req/s) | Count Latency (p99) 1min (req/s)
| Count Latency (p99) 1min (req/s) | Count 1min (errors/s)
168860566 61.76 242625.53 | 0 0 0
| 0 0 0 | 3892249 7149.31
CEP-42 with one constraint
Writes Reads
Deletes Errors
Count Latency (p99) 1min (req/s) | Count Latency (p99) 1min (req/s)
| Count Latency (p99) 1min (req/s) | Count 1min (errors/s)
163478855 57.77 246181.52 | 0 0 0
| 0 0 0 | 2784211 4237.51
164198552 57.02 245241.73 | 0 0 0
| 0 0 0 | 2784211 3898.7
165023798 55.21 245241.73 | 0 0 0
| 0 0 0 | 2812699 3898.7
165779685 55.21 246486.58 | 0 0 0
| 0 0 0 | 2812699 4042.53
166520974 55.21 246627.21 | 0 0 0
| 0 0 0 | 2812699 3719.31
167163814 57.02 246627.21 | 0 0 0
| 0 0 0 | 2818468 3719.31
167829907 57.02 244257.04 | 0 0 0
| 0 0 0 | 2818468 3514.18
168538006 52.01 244257.04 | 0 0 0
| 0 0 0 | 2852883 3514.18
169372428 52.37 244556.56 | 0 0 0
| 0 0 0 | 2875451 4144.42
170086354 54.93 244803.23 | 0 0 0
| 0 0 0 | 2875738 3817.64
Writes Reads
Deletes Errors
Count Latency (p99) 1min (req/s) | Count Latency (p99) 1min (req/s)
| Count Latency (p99) 1min (req/s) | Count 1min (errors/s)
170095971 54.93 244803.23 | 0 0 0
| 0 0 0 | 2875738 3817.64
{code}
These are both tests run against a single Cassandra process running locally on
my laptop.
> 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
>
> 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]