mirror of
https://github.com/kvcache-ai/Mooncake.git
synced 2026-11-05 01:17:55 +00:00
[PR #1261] [CLOSED] [bugfix] Metrics modifications: Deleting and adding some metrics #2114
Labels
No labels
ascend
auto-closed
bug
bug
CI/Build
Common
documentation
enhancement
feature request
good first issue
help wanted
high priority
MetaX
nightly-failure
pull-request
PyTorch Backend
question
RFC
Roadmap
stale
Store
Tests
Transfer Engine
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
kvcache-ai/Mooncake#2114
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
📋 Pull Request Information
Original PR: https://github.com/kvcache-ai/Mooncake/pull/1261
Author: @Liziqi-77
Created: 12/23/2025
Status: ❌ Closed
Base:
main← Head:main📝 Commits (5)
e7a37d8fix metrics17aa975Delete invalid coded24d99dDelete invalid codee2e4562fix formatddf2645fix format📊 Changes
9 files changed (+92 additions, -278 deletions)
View changed files
📝
mooncake-store/include/client_service.h(+0 -5)📝
mooncake-store/include/master_client.h(+0 -8)📝
mooncake-store/include/master_metric_manager.h(+13 -43)📝
mooncake-store/include/rpc_service.h(+0 -3)📝
mooncake-store/src/master_client.cpp(+0 -11)📝
mooncake-store/src/master_metric_manager.cpp(+28 -101)📝
mooncake-store/src/master_service.cpp(+20 -35)📝
mooncake-store/src/rpc_service.cpp(+0 -5)📝
mooncake-store/tests/master_metrics_test.cpp(+31 -67)📄 Description
Description
As mentioned in duplicate counts for mem_cache_hits and ssd_cache_hits, the current metrics have some issues, and this update will fix those issues.
Main modifications:
Since Mooncake can only perceive the key and not the request, it cannot calculate the true cache hit rate. Therefore, calculating it within the Conductor component is the most reasonable approach. The newly added metrics show the current usage rate of keys in the cache pool, and the update logic is as follows:
Type of Change
How Has This Been Tested?
Checklist
🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.