OpWeave: Flexible Operator Disaggregation for Heterogeneous LLM Serving

arXiv:2609.14237v1 Announce Type: cross Abstract: LLM serving systems increasingly disaggregate inference into finer-grained stages, with recent approaches separating attention from FFN or MoE execution during decode. This operator-level disaggregated serving (ODS) can improve hardware matching and enable independent scaling, particularly across heterogeneous devices. However, existing systems fix operator boundaries and lack a unified characterization of when disaggregation reduces serving cost. We present OpWeave, an end-to-end framework for heterogeneous ODS. OpWeave provides an analytical cost model that bounds the gains of homogeneous and heterogeneous ODS over colocated serving. It jointly optimizes operator partitioning and deployment configuration through a regularity-aware planner that keeps the search tractable even for hybrid-attention models. A vLLM-based runtime executes the synthesized plans with flexible operator stages across heterogeneous device groups. In our evaluation, OpWeave reduces serving cost by up to 1.78imes on homogeneous and 1.89imes on heterogeneous GPU clusters relative to the best feasible baseline, while meeting latency SLOs.

Related

• An Efficient Hybrid Sparse Attention with CPU-GPU Parallelism for Long-Context Inference
• GEM: GPU-Variability-Aware Expert to GPU Mapping for MoE Systems
• Towards Multi-Model LLM Schedulers: Empirical Insights into Offloading and Preemption

Source: arXiv cs.AI | 2026-09-15