{"ID":2848942,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.24113","arxiv_id":"2510.24113","title":"Taming the Tail: NoI Topology Synthesis for Mixed DL Workloads on Chiplet-Based Accelerators","abstract":"Heterogeneous chiplet-based systems improve scaling by disag-gregating CPUs/GPUs and emerging technologies (HBM/DRAM).However this on-package disaggregation introduces a latency inNetwork-on-Interposer(NoI). We observe that in modern large-modelinference, parameters and activations routinely move backand forth from HBM/DRAM, injecting large, bursty flows into theinterposer. These memory-driven transfers inflate tail latency andviolate Service Level Agreements (SLAs) across k-ary n-cube base-line NoI topologies. To address this gap we introduce an InterferenceScore (IS) that quantifies worst-case slowdown under contention.We then formulate NoI synthesis as a multi-objective optimization(MOO) problem. We develop PARL (Partition-Aware ReinforcementLearner), a topology generator that balances throughput, latency,and power. PARL-generated topologies reduce contention at the memory cut, meet SLAs, and cut worst-case slowdown to 1.2 times while maintaining competitive mean throughput relative to link-rich meshes. Overall, this reframes NoI design for heterogeneouschiplet accelerators with workload-aware objectives.","short_abstract":"Heterogeneous chiplet-based systems improve scaling by disag-gregating CPUs/GPUs and emerging technologies (HBM/DRAM).However this on-package disaggregation introduces a latency inNetwork-on-Interposer(NoI). We observe that in modern large-modelinference, parameters and activations routinely move backand forth from HBM...","url_abs":"https://arxiv.org/abs/2510.24113","url_pdf":"https://arxiv.org/pdf/2510.24113v1","authors":"[\"Arnav Shukla\",\"Harsh Sharma\",\"Srikant Bharadwaj\",\"Vinayak Abrol\",\"Sujay Deb\"]","published":"2025-10-28T06:36:44Z","proceeding":"cs.AR","tasks":"[\"cs.AR\",\"cs.AI\",\"cs.LG\"]","methods":"[]","has_code":false}