[GitHub] [spark] Ngone51 opened a new pull request #25577: [WIP][CORE][SPARK-28867] InMemoryStore checkpoint to speed up replay log file in HistoryServer

[GitBox <gi...@apache.org>]

Ngone51 opened a new pull request #25577: [WIP][CORE][SPARK-28867] InMemoryStore checkpoint to speed up replay log file in HistoryServer
URL: https://github.com/apache/spark/pull/25577
 
 
   <!--
   Thanks for sending a pull request!  Here are some tips for you:
     1. If this is your first time, please read our contributor guidelines: https://spark.apache.org/contributing.html
     2. Ensure you have added or run the appropriate tests for your PR: https://spark.apache.org/developer-tools.html
     3. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP][SPARK-XXXX] Your PR title ...'.
     4. Be sure to keep the PR description updated to reflect all changes.
     5. Please write your PR title to summarize what this PR proposes.
     6. If possible, provide a concise example to reproduce the issue for a faster review.
   -->
   
   ### What changes were proposed in this pull request?
   <!--
   Please clarify what changes you are proposing. The purpose of this section is to outline the changes and how this PR fixes the issue. 
   If possible, please consider writing useful notes for better and faster reviews in your PR. See the examples below.
     1. If you refactor some codes with changing classes, showing the class hierarchy will help reviewers.
     2. If you fix some SQL features, you can provide some references of other DBMSes.
     3. If there is design documentation, please add the link.
     4. If there is a discussion in the mailing list, please add the link.
   -->
   
   This PR aims to improve the replay performance in HistoryServer by periodically checkpoint InMemoryStore in an in-completed application and achieve incremental replay.The main idea
   is, for an in-completed application, we periodically (normally every N events num) checkpoint InMemoryStore with processed events num(X) into event log dir. And in HistoryServer, it reconstructs InMemoryStore from checkpoint file and gets X. Then, we could skip X events while replaying the log file basing on the partial InMemoryStore. Note that we should also recover those live entities from the the partial InMemoryStore in AppStatusListener to perform incremental replay. For a completed application, HistoryServer could just reconstructs InMemoryStore and no need to do replay.
   
   And in this PR, we only focus on handling InMemoryStore in HistoryServer, while LevelDB is planed to be handled in similar way in following PR.
   
   Basic experiment on a completed application of 20055 events shows the improvement of this optimization:
   
   without optimization  | with optimization(including deserialization time)
    :-: | :-: | 
   4343 | 78(70)
   4512 | 92(85)
   4475 | 74(68)
   4254 | 93(78)
   4126 | 81(71)
   
   Work TODO
   
   - [ ] compression support when checkpoint InMemoryStore 
   - [ ] More accurate conversion from wrapper data to live entity
   - [ ] checkpoint file cleaning in HistoryServer
   - [ ] overcome frequently StackOverError in deserialization
   - [ ] unit tests
   
   ### Why are the changes needed?
   <!--
   Please clarify why the changes are needed. For instance,
     1. If you propose a new API, clarify the use case for a new API.
     2. If you fix a bug, you can clarify why it is a bug.
   -->
   
   Change is needed because HistoryServer now could be very slow to replay a large log file at the first time and it always re-replay an in-progress log file after it changes which leads to low efficiency. 
   
   ### Does this PR introduce any user-facing change?
   <!--
   If yes, please clarify the previous behavior and the change this PR proposes - provide the console output, description and/or an example to show the behavior difference if possible.
   If no, write 'No'.
   -->
   
   Yes, if user wants to use this optimization by several new configurations.  
   
   ### How was this patch tested?
   <!--
   If tests were added, say they were added here. Please make sure to add some test cases that check the changes thoroughly including negative and positive cases if possible.
   If it was tested in a way different from regular unit tests, please clarify how you tested step by step, ideally copy and paste-able, so that other reviewers can test and check, and descendants can verify in the future.
   If tests were not added, please describe why they were not added and/or why it was difficult to add.
   -->
   
   Only tested manually yet, still work in process.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org