store/copr: set low concurrency for keep order coprocessor requests

[tiancaiamao]

What problem does this PR solve?

Issue Number: close #35974

Problem Summary:

keep order requests consume much more memory than normal cases.
That's because we use a high coprocessor concurrency for it, but with the existance of 'order by',
the cop responses have to be handled one by one ... most of the cop responses are just wasting memory.

What is changed and how it works?

We don't need high concurrency for keep order coprocessor requests.

time mysql -h 127.0.0.1 -u root -P 4000 -B ens_rb -e 'set @@tidb_enable_paging = 1; select * from ens_rb.fm_acct_client_relation order by loginid' > /dev/null

=====================

The performance do not decrease with lower concurrency!

Before:

real	2m30.225s
user	1m33.578s
sys	0m8.454s

After:

real	2m32.266s
user	1m35.731s
sys	0m7.367s

=====================

The memory consumption reduced:

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

Please refer to Release Notes Language Style Guide to write a quality release note.

Set a low concurrency value for coprocessor requests with 'keep order' property, this change can dramatically reduce the memory usage in a full table scan with 'order by' scenario 

[ti-chi-bot]

[REVIEW NOTIFICATION]

This pull request has been approved by:

• XuHuaiyu
• wshwsh12

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.

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

[tiancaiamao]

Testing with non-paging protocol, the conclusion is the same, high concurrency is useless here.

time mysql -h [127.0.0.1](http://127.0.0.1/) -u root -P 4000 -B ens_rb -e 'set @@tidb_enable_paging = 0; select * from ens_rb.fm_acct_client_relation order by loginid' > /dev/null

real	2m28.491s
user	1m35.039s
sys	0m7.399s

time mysql -h [127.0.0.1](http://127.0.0.1/) -u root -P 4000 -B ens_rb -e 'set @@tidb_enable_paging = 0; select * from ens_rb.fm_acct_client_relation order by loginid' > /dev/null

real	2m28.764s
user	1m35.245s
sys	0m7.526s

[wshwsh12]

LGTM

[tiancaiamao]

/merge

[ti-chi-bot]

This pull request has been accepted and is ready to merge.

Commit hash: 8f6ab49

[ti-chi-bot]

@tiancaiamao: 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

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.

[tiancaiamao]

/run-unit-test

[tiancaiamao]

/hold

[tiancaiamao]

/run-unit-test

[sre-bot]

Code Coverage Details: https://codecov.io/github/pingcap/tidb/commit/74edaebe3e418a34515791a805320c2bc488204b

[tiancaiamao]

/run-check_dev_2

[tiancaiamao]

/run-check_dev_2

[tiancaiamao]

/unhold

[tiancaiamao]

/merge

[ti-chi-bot]

This pull request has been accepted and is ready to merge.

Commit hash: 74edaeb