← 研踪 PaperTrailarXiv:2611.23883 · v1
Accelerating Distributed Digital Ising Machines
首次提交 2026-11-18 13:35
版本更新 2026-11-18 13:35
雷达观测 2026-11-19 19:08
摘要
Ising machines, dedicated accelerators built around the dynamics of Ising-like spin models, have emerged as a leading platform for combinatorial optimization and probabilistic sampling. The problems that motivate them require millions of spins, far beyond the on-chip capacity of any single device, so a practical Ising machine must distribute its computation across many devices. Prior work reports throughput at a single operating point, leaving the binding resource unidentified and platform crossovers measurable only by experiment; no calibrated cross-platform cost model of distributed Ising machines has been reported, nor have systematic techniques to accelerate their Markov chain Monte Carlo (MCMC) workloads. This work presents a calibrated cost model that decomposes digital Ising sweep time into compute, memory, synchronization, and inter-device communication terms, validated with 4.8% mean error across three GPU generations, FPGA and ASIC up to 1 million spins. Guided by the model, we address each term in turn and propose: (i) a data placement strategy, worth up to 4.2 speedup; (ii) multi-rate and bounded-staleness schedules that relax the barrier that Gibbs sampling imposes, cutting time-to-solution (TTS) by up to 3.16 on three benchmarks; and (iii) a phase-offset streaming dataflow that hides boundary exchange behind interior computation whenever a link satisfies a latency-transparency criterion, yielding up to 1.25/1.74/2.01 on an H200 pair, a ZCU106 pair, and a four-board XEM8320 mesh. The streaming dataflow is also implemented on a fabricated four-chip 28 nm ASIC sustaining 28.1 Gflips/s at 4.81 pJ/flip.
作者备注:16 pages. Accepted at HPCA
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-18 13:35 · 最新观测版本
Accelerating Distributed Digital Ising Machines
Ising machines, dedicated accelerators built around the dynamics of Ising-like spin models, have emerged as a leading platform for combinatorial optimization and probabilistic sampling. The problems that motivate them require millions of spins, far beyond the on-chip capacity of any single device, so a practical Ising machine must distribute its computation across many devices. Prior work reports throughput at a single operating point, leaving the binding resource unidentified and platform crossovers measurable only by experiment; no calibrated cross-platform cost model of distributed Ising machines has been reported, nor have systematic techniques to accelerate their Markov chain Monte Carlo (MCMC) workloads. This work presents a calibrated cost model that decomposes digital Ising sweep time into compute, memory, synchronization, and inter-device communication terms, validated with 4.8% mean error across three GPU generations, FPGA and ASIC up to 1 million spins. Guided by the model, we address each term in turn and propose: (i) a data placement strategy, worth up to 4.2 speedup; (ii) multi-rate and bounded-staleness schedules that relax the barrier that Gibbs sampling imposes, cutting time-to-solution (TTS) by up to 3.16 on three benchmarks; and (iii) a phase-offset streaming dataflow that hides boundary exchange behind interior computation whenever a link satisfies a latency-transparency criterion, yielding up to 1.25/1.74/2.01 on an H200 pair, a ZCU106 pair, and a four-board XEM8320 mesh. The streaming dataflow is also implemented on a fabricated four-chip 28 nm ASIC sustaining 28.1 Gflips/s at 4.81 pJ/flip.
查看此版本解读 →查看此版本原文 ↗