Storages: Refine SegmentReadTaskPool and SegmentReadTaskScheduler

[JinheLin]

What problem does this PR solve?

Issue Number: ref #6834

What is changed and how it works?

• This PR mainly refines SegmentReadTaskPool and SegmentReadTaskScheduler preparing for supporting read-thread and data-sharing in disaggregated-mode. The logic of the reading path is still not changed after this PR.

Main changes:

• Remove DMContext and physical_table_id from SegmentReadTaskPool.
• Add DMContext to SegmentReadTask.
  • In disaggregated-mode, EstablishDisaggTask can return SegmentReadTasks of different physical tables. So SegmentReadTaskPool needs to support SegmentReadTasks of different physical tables.
  • Move information of physical tables in DMContext into SegmentReadTask .
• Use GlobalSegmentID to identify a segment in SegmentReadTaskScheduler.
  • In op-mode, physical_table_id + segment_id + segment_epoch can identify a segment. In disaggregated-mode, it also needs to add store_id and keyspace_id.
  • Wrap this information into struct GlobalSegmentID.
• Enhance the formatter of SegmentReadTaskPtr and remove SegmentReadTask::info.
• Remove cluster member from RNWorkerFetchPages.
• Remove RNReadTask since it is redundant.

Check List

Tests

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

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

None

[ti-chi-bot]

Skipping CI for Draft Pull Request.
If you want CI signal for your change, please convert it to an actual PR.
You can still manually trigger a test run with /test all

[JinheLin]

/run-all-tests

[JinheLin]

/run-all-tests

[JaySon-Huang]

Rest LGTM

[JaySon-Huang]

I'd like to keep a cluster in RNWorkerFetchPages. Getting it from seg_task->dm_context->db_context.... is too implicit and hard to write tests (if we do write tests later)

[JinheLin]

The logic of the fetchpages is not pure, it depends on other member from dm_context/db_context. We cannot remove dm_context/db_context from fetchpages easily, so keep cluster member seems not make much sense.

And I think the main logic of fetchpages should be member functions of SegmentReadTask, instead of part the thread workers. The code also needs to be refined later.

If we write tests for it , we'd better to refine functions to accept a cluster argument or set the member of dm_context properly.

[JaySon-Huang]

OK

[JaySon-Huang]

LGTM

[JinheLin]

/run-unit-test

[JaySon-Huang]

/run-unit-test

[ti-chi-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: JaySon-Huang, Lloyd-Pottiger

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

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [JaySon-Huang,Lloyd-Pottiger]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[ti-chi-bot]

[LGTM Timeline notifier]

Timeline:

• 2023-10-27 10:05:37.993442257 +0000 UTC m=+2601935.580552401: ☑️ agreed by JaySon-Huang.
• 2023-10-28 00:48:16.548265353 +0000 UTC m=+2654894.135375498: ☑️ agreed by Lloyd-Pottiger.

[ti-chi-bot]

@JinheLin: Your PR was out of date, I have automatically updated it for you.

At the same time I will also trigger all tests for you:

/run-all-tests

trigger some heavy tests which will not run always when PR updated.

If the CI test fails, you just re-trigger the test that failed and the bot will merge the PR for you after the CI passes.

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 ti-community-infra/tichi repository.

[JinheLin]

/run-all-tests