PageStorage: Fix peak memory usage when running GC on PageDirectory (#6168)

[ti-chi-bot]

This is an automated cherry-pick of #6168

What problem does this PR solve?

Issue Number: close #6163

Problem Summary:

The snapshot file getting bigger and bigger, causing peak memory usage after TiFlash running for a long time.
Introduced by #5357

#5357 want to resolve a bug that cause by concurrency between deleting a page and Full GC. But that fixes will make the entries in PageDirectory snapshot file getting bigger and bigger. Cause it can not ensure whether there are more edit on the pages with a FileSnapshot

What is changed and how it works?

TBD

Check List

Tests

• Unit test
• Integration test
• Manual test (add detailed scripts or steps below)
• No code

Run the HeavySkewWriteRead testing pageworkload (based on #6178), we can see that before this PR, the wal snapshot file getting bigger and bigger as it does not remove the deleted page from snapshot file. After this PR, the wal snapshot file only keep the available page.

Before this PR

... ["Dumped directory snapshot ... [num_records=1085673] ... [size=76022852]."] [source=stress_test] [thread_id=23]
... ["Dumped directory snapshot ... [num_records=1125073] ... [size=78780852]."] [source=stress_test] [thread_id=23]
... ["Dumped directory snapshot ... [num_records=1168805] ... [size=81842092]."] [source=stress_test] [thread_id=23]
... ["Dumped directory snapshot ... [num_records=1213833] ... [size=84994052]."] [source=stress_test] [thread_id=23]
... ["Dumped directory snapshot ... [num_records=1258217] ... [size=88100932]."] [source=stress_test] [thread_id=23]
... ["Dumped directory snapshot ... [num_records=1307151] ... [size=91526312]."] [source=stress_test] [thread_id=23]

After this PR

... ["Dumped directory snapshot ... [num_records=754] ... [size=78420]."] [source=stress_test] [thread_id=24]
... ["Dumped directory snapshot ... [num_records=754] ... [size=78420]."] [source=stress_test] [thread_id=24]
... ["Dumped directory snapshot ... [num_records=755] ... [size=78524]."] [source=stress_test] [thread_id=24]
... ["Dumped directory snapshot ... [num_records=754] ... [size=78420]."] [source=stress_test] [thread_id=24]
... ["Dumped directory snapshot ... [num_records=754] ... [size=78420]."] [source=stress_test] [thread_id=24]

Side effects

• Performance regression: Consumes more CPU
• Performance regression: Consumes more Memory
• Breaking backward compatibility

Documentation

• Affects user behaviors
• Contains syntax changes
• Contains variable changes
• Contains experimental features
• Changes MySQL compatibility

Release note

Fix the bug that leads to peak memory usage when running GC

[ti-chi-bot]

[REVIEW NOTIFICATION]

This pull request has not been approved.

To complete the pull request process, please ask the reviewers in the list to review by filling /cc @reviewer in the comment.
After your PR has acquired the required number of LGTMs, you can assign this pull request to the committer in the list by filling /assign @committer in the comment to help you merge this pull request.

The full list of commands accepted by this bot can be found here.

Details

Reviewer can indicate their review by submitting an approval review.
Reviewer can cancel approval by submitting a request changes review.

[ti-chi-bot]

@ti-chi-bot: This cherry pick PR is for a release branch and has not yet been approved by release team.
Adding the do-not-merge/cherry-pick-not-approved label.

To merge this cherry pick, it must first be approved by the collaborators.

AFTER it has been approved by collaborators, please ping the release team in a comment to request a cherry pick review.

Details

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.