apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-06 00:57:33 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[optimize] Optimize prefix caching in v1 release/2.1 (#3823)

Browse Source 
* [optimize] Optimize prefix caching in v1

* [optimize] Optimize prefix caching in v1

---------

Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>
This commit is contained in:
chenjian
2025-09-04 19:25:02 +08:00
committed by GitHub
parent c2f5c99b1e
commit ffec66097c
3 changed files with 49 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fastdeploy/engine/engine.py
View File

@@ -355,9 +355,9 @@ class LLMEngine:
self.cfg.max_prefill_batch,
)
self.resource_manager.check_and_free_block_tables()
# self.resource_manager.check_and_free_block_tables()
tasks = self.scheduler.get_requests(
available_blocks=self.resource_manager.available_block_num(),
available_blocks=self.cfg.cache_config.max_block_num_per_seq,
block_size=self.cfg.cache_config.block_size,
reserved_output_blocks=self.cfg.cache_config.enc_dec_block_num,
max_num_batched_tokens=self.cfg.max_model_len,