Skip to content

Support l2 adapter check and improve basic_check tool #2895

Merged
chunxiaozheng merged 5 commits intoLMCache:devfrom
maobaolong:check_l2_adapter
Mar 31, 2026
Merged

Support l2 adapter check and improve basic_check tool #2895
chunxiaozheng merged 5 commits intoLMCache:devfrom
maobaolong:check_l2_adapter

Conversation

@maobaolong
Copy link
Copy Markdown
Collaborator

@maobaolong maobaolong commented Mar 28, 2026

What this PR does / why we need it:

introduces a new test mode for L2 adapters and enhances existing remote and storage manager test modes by adding support for configurable object sizes, KV data types, and settle times. It also updates the performance reporting utilities to calculate and display throughput

Special notes for your reviewers:

If applicable:

  • this PR contains user facing changes - docs added
  • this PR contains unit tests

Signed-off-by: baoloongmao <baoloongmao@tencent.com>
Signed-off-by: baoloongmao <baoloongmao@tencent.com>
Signed-off-by: baoloongmao <baoloongmao@tencent.com>
@maobaolong maobaolong requested a review from sammshen March 28, 2026 00:21
Copy link
Copy Markdown
Contributor

@gemini-code-assist gemini-code-assist Bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request introduces a new test mode for L2 adapters and enhances existing remote and storage manager test modes by adding support for configurable object sizes, KV data types, and settle times. It also updates the performance reporting utilities to calculate and display throughput. A logic issue was identified in the new L2 adapter test script where the timing variable for the initial lookup phase is overwritten by a subsequent phase, leading to inaccurate performance reporting in the final summary.

Comment thread lmcache/v1/check/check_mode_test_l2_adapter.py Outdated
Signed-off-by: baoloongmao <baoloongmao@tencent.com>
Signed-off-by: baoloongmao <baoloongmao@tencent.com>
Copy link
Copy Markdown
Contributor

@sammshen sammshen left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM!

@deng451e deng451e self-requested a review March 31, 2026 01:27
Copy link
Copy Markdown
Collaborator

@deng451e deng451e left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@maobaolong
Copy link
Copy Markdown
Collaborator Author

@chunxiaozheng Would you like to help to check this PR works for the internal backend?

Copy link
Copy Markdown
Collaborator

@chunxiaozheng chunxiaozheng left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM!

@chunxiaozheng chunxiaozheng enabled auto-merge (squash) March 31, 2026 11:53
@github-actions github-actions Bot added the full Run comprehensive tests on this PR label Mar 31, 2026
@chunxiaozheng chunxiaozheng merged commit 3ac1499 into LMCache:dev Mar 31, 2026
44 of 46 checks passed
jooho-XCENA pushed a commit to xcena-dev/LMCache that referenced this pull request Apr 2, 2026
* Support l2 adapter check

Signed-off-by: baoloongmao <baoloongmao@tencent.com>

* Support display throughput and check specify num keys

Signed-off-by: baoloongmao <baoloongmao@tencent.com>

* Add settle time

Signed-off-by: baoloongmao <baoloongmao@tencent.com>

* fix: use separate vars for absent/exist lookup timing

Signed-off-by: baoloongmao <baoloongmao@tencent.com>

* Fix ut

Signed-off-by: baoloongmao <baoloongmao@tencent.com>

---------

Signed-off-by: baoloongmao <baoloongmao@tencent.com>
jooho-XCENA pushed a commit to xcena-dev/LMCache that referenced this pull request Apr 2, 2026
* Support l2 adapter check

Signed-off-by: baoloongmao <baoloongmao@tencent.com>

* Support display throughput and check specify num keys

Signed-off-by: baoloongmao <baoloongmao@tencent.com>

* Add settle time

Signed-off-by: baoloongmao <baoloongmao@tencent.com>

* fix: use separate vars for absent/exist lookup timing

Signed-off-by: baoloongmao <baoloongmao@tencent.com>

* Fix ut

Signed-off-by: baoloongmao <baoloongmao@tencent.com>

---------

Signed-off-by: baoloongmao <baoloongmao@tencent.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

full Run comprehensive tests on this PR

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants