[BugFix] Refine the preparation of cpu and storage cache

[juncaipeng]

Motivation

统一准备cpu和storage的cache

Modifications

prefix-cache-manager模块

Usage or Command

不变

Accuracy Tests

后续更新ci镜像后添加单侧

Checklist

• Add at least a tag in the PR title.
  • Tag list: [[FDConfig],[APIServer],[Engine], [Scheduler], [PD Disaggregation], [Executor], [Graph Optimization], [Speculative Decoding], [RL], [Models], [Quantization], [Loader], [OP], [KVCache], [DataProcessor], [BugFix], [Docs], [CI], [Optimization], [Feature], [Benchmark], [Others], [XPU], [HPU], [GCU], [DCU], [Iluvatar], [Metax]]
  • You can add new tags based on the PR content, but the semantics must be clear.
• Format your code, run pre-commit before commit.
• Add unit tests. Please write the reason in this PR if no unit tests.
• Provide accuracy results.
• If the current PR is submitting to the release branch, make sure the PR has been submitted to the develop branch, then cherry-pick it to the release branch with the [Cherry-Pick] PR tag.

[paddle-bot]

Thanks for your contribution!

[Copilot]

Pull request overview

This PR refactors the cache preparation logic to unify the handling of CPU and storage cache within the prefix_cache_manager module. The main motivation is to consolidate scattered cache preparation code and eliminate duplication between CPU and storage cache handling.

Key Changes:

• Integrated storage cache matching and preparation logic into the request_match_blocks method in prefix_cache_manager
• Removed the separate get_storage_cached_blocks method from resource_manager_v1 (though calls to it remain in some code paths outside this diff)
• Split the gpu_cpu_cache_prepare_time metric into separate cpu_cache_prepare_time and storage_cache_prepare_time metrics for better observability

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
fastdeploy/engine/sched/resource_manager_v1.py	Removed duplicate storage cache logic and unused import; updated get_prefix_cached_blocks to handle new consolidated metrics
fastdeploy/cache_manager/prefix_cache_manager.py	Integrated storage cache matching into request_match_blocks; added storage cache preparation logic with proper block allocation and recycling; updated return value of issue_prefetch_storage_task to return count instead of list
fastdeploy/engine/request.py	Renamed metric field from gpu_cpu_cache_prepare_time to cpu_cache_prepare_time to better reflect its purpose
fastdeploy/cache_manager/cache_metrics.py	Added storage cache token tracking throughout metrics calculation and reporting
fastdeploy/cache_manager/cache_transfer_manager.py	Improved debug logging by showing counts instead of full lists for better performance
benchmarks/benchmark_serving.py	Updated to use new cpu_cache_prepare_time metric name
benchmarks/backend_request_func.py	Updated to use new cpu_cache_prepare_time metric name

[codecov-commenter]

Codecov Report

❌ Patch coverage is 43.52941% with 48 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@9018ccf). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
fastdeploy/cache_manager/prefix_cache_manager.py	50.00%	24 Missing and 3 partials ⚠️
fastdeploy/engine/sched/resource_manager_v1.py	8.33%	11 Missing ⚠️
fastdeploy/cache_manager/cache_transfer_manager.py	0.00%	10 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             develop    #5777   +/-   ##
==========================================
  Coverage           ?   66.71%           
==========================================
  Files              ?      347           
  Lines              ?    44354           
  Branches           ?     6810           
==========================================
  Hits               ?    29591           
  Misses             ?    12578           
  Partials           ?     2185           

Flag	Coverage Δ
GPU	66.71% <43.52%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Copilot]

Pull request overview

Copilot reviewed 10 out of 10 changed files in this pull request and generated 3 comments.