You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Hangxiang Yu (Jira)" <ji...@apache.org> on 2023/11/20 07:17:00 UTC

[jira] [Resolved] (FLINK-30235) Comprehensive benchmarks on changelog checkpointing

     [ https://issues.apache.org/jira/browse/FLINK-30235?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Hangxiang Yu resolved FLINK-30235.
----------------------------------
    Fix Version/s: 1.17.0
         Assignee: Rui Xia
       Resolution: Fixed

Resolved this in 1.17.0
See theĀ [blog post|https://www.ververica.com/blog/generic-log-based-incremental-checkpoint] for more details about the benchmark results.

> Comprehensive benchmarks on changelog checkpointing
> ---------------------------------------------------
>
>                 Key: FLINK-30235
>                 URL: https://issues.apache.org/jira/browse/FLINK-30235
>             Project: Flink
>          Issue Type: Improvement
>          Components: Runtime / Checkpointing, Runtime / State Backends
>            Reporter: Rui Xia
>            Assignee: Rui Xia
>            Priority: Minor
>              Labels: performance
>             Fix For: 1.17.0
>
>
> Changelog checkpointing is functionally usable right now. To make it as a productive feature, more comprehensive benchmarks are required. In this issue, I aim to answer the following two major concerns:
>  * The expansion of full checkpoint size caused by changelog persistence;
>  * The TPS regression caused by DTSL double-write;
> By the way, I will also present other metrics related to checkpointing.



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