[fix] [broker] remove usage of deprecated loadBalancerMemoryResourceWeight.

[thetumbled]

Motivation

loadBalancerMemoryResourceWeight has been deprecated in #19559.
But it still work at some logic.

Modifications

Remove it completely.

Verifying this change

• Make sure that the change passes the CI checks.

(Please pick either of the following options)

This change is a trivial rework / code cleanup without any test coverage.

Does this pull request potentially affect one of the following parts:

If the box was checked, please highlight the changes

• Dependencies (add or upgrade a dependency)
• The public API
• The schema
• The default values of configurations
• The threading model
• The binary protocol
• The REST endpoints
• The admin CLI options
• The metrics
• Anything that affects deployment

Documentation

• doc
• doc-required
• doc-not-needed
• doc-complete

Matching PR in forked repository

PR in forked repository: thetumbled#55

[thetumbled]

PTAL, thanks. @lhotari @dao-jun @Demogorgon314 @heesung-sn

[lhotari]

@thetumbled there's #21973 describing a way how to fix the issue in estimating the heap memory usage. if that is done, it would be possible to keep on using it. It's a major gap if the heap memory usage isn't considered at all.

[thetumbled]

@thetumbled there's #21973 describing a way how to fix the issue in estimating the heap memory usage. if that is done, it would be possible to keep on using it. It's a major gap if the heap memory usage isn't considered at all.

Do we have a plan for using this configuration again? If #21973 enhance the collection of memory usage, will memory usage have a strong relation with the actual load of the broker?
WDYT, @heesung-sn @Demogorgon314 @codelipenghui @Technoboy-

[thetumbled]

I start a discussion for this pr: https://lists.apache.org/thread/qpdvyytk8y7x8yzh35jshksnl6498y3r, looking forward to your feedback.
And this is the implementation PR for this thread: #22820

[heesung-sn]

Removing the memory usage signals in the code is probably a radical decision here. We could disable them by default, but we could still let users use them if they want to. #21973 seems promising to improve the memory signals in the future.

[thetumbled]

Removing the memory usage signals in the code is probably a radical decision here. We could disable them by default, but we could still let users use them if they want to. #21973 seems promising to improve the memory signals in the future.

Yes, PR #22820 is to disable them by default. PTAL, thanks.

[thetumbled]

As #22820 has been merged, i close this pr and looking forward to the improvement of memory usage collection.
If needed, we can still reopen this pr to remove the logic about memory usage thoroughly.