LMetric routing policy (OSDI'26) + A/B results vs linear baseline
Implement LMetric (P_tokens × BS multiplication score) from "Simple is
Better" (Zhang et al., OSDI'26) as alternative routing policy for
combined mode. Key changes:
- cache_aware_proxy.py: add --policy {linear,lmetric} flag, track
pending_prefill_tokens and num_requests per instance, /stats endpoint
- run_lmetric_ab.sh: automated A/B script for fair comparison
Results (200 req, fresh restart, same trace):
Linear: TTFT50=1.086 TPOT90=0.077 E2E50=5.423
LMetric: TTFT50=1.099 TPOT90=0.073 E2E50=5.205
Delta: TTFT +1.2% TPOT -5.9% E2E -4.0%
LMetric improves TPOT/E2E modestly through better load balancing, but
routing policy headroom is limited vs elastic P2P offload (-44% E2E).
TODO: vLLM → Redis → router pipeline for exact state ablation.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
5
TODO.md
5
TODO.md
@@ -21,5 +21,10 @@ GPU 机器:dash0,是 8*H20 的机器,可以直接 `ssh dash0` 进行连接
|
||||
3. 跑通 PD 分离,确认 PD 分离能够比普通的 PD 混合一起跑的性能要好,给出两者详细的性能对比以及原因分析
|
||||
4. 判断 trace 的 pattern,是否有必要 PD 完全混合或者 PD 分离
|
||||
5. 参考本地的 `~/phd/agentic-pd-hybrid`,判断是否能够实现一套 prefill-as-a-service 的架构,把重的 prefill 交给 prefill service,prefill service 能够从本地 GPU/DRAM/别的 GPU 机器上 pull KVCache,提高本地的 prefix KVCache hit ratio,不影响 decoding 的 prefill,就可以交给过去 PD 分离定义中 D-node 来做,提高 KVCache 命中率
|
||||
6. [TODO] Router 侧状态精确性 ablation:当前 router 自己维护 shadow state(ongoing_tokens, pending_prefill_tokens, cached_blocks 等),与 vLLM 引擎内部真实状态可能存在 gap(尤其是 cache eviction 导致的 APC 偏差)。需要:
|
||||
- 修改 vLLM,让每个 instance 定期把内部状态(waiting queue depth, running batch size, KV cache usage, actual prefix cache hit blocks)写入 Redis
|
||||
- Router 从 Redis 读取精确状态,替代当前的 proxy-side 近似
|
||||
- 对比 proxy shadow state vs Redis exact state 的 routing 决策差异和最终性能差异
|
||||
- 量化 gap 大小:哪些指标差距最大?是否影响实际 routing 质量?
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user