20260420:first commit

raw/papers/qin-prfaas-cross-datacenter-2026.md

@@ -0,0 +1,24 @@
+---
+title: "Prefill-as-a-Service: KVCache of Next-Generation Models Could Go Cross-Datacenter"
+arxiv_id: "2604.15039"
+authors: ["Ruoyu Qin", "Weiran He", "Yaoyu Wang", "Zheming Li", "Xinran Xu", "Yongwei Wu", "Weimin Zheng", "Mingxing Zhang"]
+published: "2026-04-16"
+updated: "2026-04-16"
+categories: ["cs.DC"]
+primary_category: "cs.DC"
+url: "https://arxiv.org/abs/2604.15039"
+abstract: |
+  Prefill-decode (PD) disaggregation has become the standard architecture for large-scale LLM serving, but in practice its deployment boundary is still determined by KVCache transfer. In conventional dense-attention models, prefill generates huge KVCache traffics that keep prefill and decode tightly coupled within a single high-bandwidth network domain, limiting heterogeneous deployment and resource elasticity. Recent hybrid-attention architectures substantially reduce KVCache size, making cross-cluster KVCache transport increasingly plausible. However, smaller KVCache alone does not make heterogeneous cross-datacenter PD serving practical: real workloads remain bursty, request lengths are highly skewed, prefix caches are unevenly distributed, and inter-cluster bandwidth fluctuates. A naive design that fully externalizes prefill can therefore still suffer from congestion, unstable queueing, and poor utilization.
+  We present Prefill-as-a-Service (PrfaaS), a cross-datacenter serving architecture that selectively offloads long-context prefill to standalone, compute-dense prefill clusters and transfers the resulting KVCache over commodity Ethernet to local PD clusters for decode. Rather than treating reduced KVCache as sufficient, PrfaaS combines model-side KV efficiency with system-side selective offloading, bandwidth-aware scheduling, and cache-aware request placement. This design removes the requirement that heterogeneous accelerators share the same low-latency RDMA fabric, enabling independent scaling of prefill and decode capacity across loosely coupled clusters. In a case study using an internal 1T-parameter hybrid model, a PrfaaS-augmented heterogeneous deployment achieves 54% and 32% higher serving throughput than homogeneous PD and naive heterogeneous baselines, respectively, while consuming only modest cross-datacenter bandwidth.
+---
+
+# Prefill-as-a-Service: KVCache of Next-Generation Models Could Go Cross-Datacenter
+
+**arXiv:** 2604.15039 [cs.DC]
+**Published:** 2026-04-16
+**Authors:** Ruoyu Qin, Weiran He, Yaoyu Wang, Zheming Li, Xinran Xu, Yongwei Wu, Weimin Zheng, Mingxing Zhang
+
+## Abstract
+
+Prefill-decode (PD) disaggregation has become the standard architecture for large-scale LLM serving, but in practice its deployment boundary is still determined by KVCache transfer. In conventional dense-attention models, prefill generates huge KVCache traffics that keep prefill and decode tightly coupled within a single high-bandwidth network domain, limiting heterogeneous deployment and resource elasticity. Recent hybrid-attention architectures substantially reduce KVCache size, making cross-cluster KVCache transport increasingly plausible. However, smaller KVCache alone does not make heterogeneous cross-datacenter PD serving practical: real workloads remain bursty, request lengths are highly skewed, prefix caches are unevenly distributed, and inter-cluster bandwidth fluctuates. A naive design that fully externalizes prefill can therefore still suffer from congestion, unstable queueing, and poor utilization.
+  We present Prefill-as-a-Service (PrfaaS), a cross-datacenter serving architecture that selectively offloads long-context prefill to standalone, compute-dense prefill clusters and transfers the resulting KVCache over commodity Ethernet to local PD clusters for decode. Rather than treating reduced KVCache as sufficient, PrfaaS combines model-side KV efficiency with system-side selective offloading, bandwidth-aware scheduling, and cache-aware request placement. This design removes the requirement that heterogeneous accelerators share the same low-latency RDMA fabric, enabling independent scaling of prefill and decode capacity across loosely coupled clusters. In a case study using an internal 1T-parameter hybrid model, a PrfaaS-augmented heterogeneous deployment achieves 54% and 32% higher serving throughput than homogeneous PD and naive heterogeneous baselines, respectively, while consuming only modest cross-datacenter bandwidth.