DFU-L: A Latency-Sensitive Dataflow Architecture with NoC and Streaming Optimizations for Edge DSP

Abstract

Dataflowarchitectures are attractive for real-time edgeworkloads because structured computation and streaming data movement expose abundant instruction-level and pipeline parallelism. However, as working sets shrink and iteration latency becomes critical, conventional spatial dataflow fabrics can become limited by on-chip transfer and control overheads rather than raw compute throughput. Fine-grained inter-PE dependences, scratchpad-bank contention, and frequent instruction or kernel switching introduce non-negligible stalls, reducing effective PE utilization. This paper presents DFU-L, a latency-sensitive coarse-grained dataflow architecture that improves latency overlap across on-chip communication, streaming memory access, and execution switching. DFU-L first introduces a forwarding NoC with dataflow-unblocking arbitration to accelerate inter-PE operand delivery and reduce dependence-transfer latency. It then organizes memory accesses into clustered fetch domains and uses a paced streaming engine to smooth scratchpad-bank service under memory-bound thin-DFG execution. Finally, DFU-L reuses idle forwarding windows to overlap instruction and kernel switching with the tail of the current dataflow execution, reducing control-transition overhead. We evaluate DFU-L with a cycle-accurate simulator calibrated against RTL and a 12 nm, 1 GHz implementation. Across representative edge DSP kernels, DFU-L achieves up to 1.92× speedup and 1.45× on average over a mesh baseline. It improves energy efficiency by 1.14×–2.06× over Plasticine, reduces energy-delay product by up to 2.18× compared with state-of-the-art DSP accelerators, and achieves 1.11–1.37× higher normalized energy efficiency than advanced dataflow designs, under analytical process-technology normalization.

Keywords

Dataflow architecture / edge DSP workloads / network-on-chip / streaming memory access / latency-sensitive execution

Rights & permissions

Higher Education Press 2026

Just Accepted

This article has successfully passed peer review and final editorial review, and will soon enter typesetting, proofreading and other publishing processes. The currently displayed version is the accepted final manuscript. The officially published version will be updated with format, DOI and citation information upon launch. We recommend that you pay attention to subsequent journal notifications and preferentially cite the officially published version. Thank you for your support and cooperation.