D-PDLP:将PDHG算法扩展到分布式多GPU系统
D-PDLP: Scaling PDLP to Distributed Multi-GPU Systems
AI总结:
本文提出D-PDLP框架,通过二维网格划分将PDHG算法扩展到多GPU环境,结合块状随机排列和非零元素意识矩阵划分,提升负载平衡和计算效率,实现在大规模线性规划问题上的高扩展性和性能。
摘要:
我们提出了一种分布式Primal-Dual Hybrid Gradient(PDHG)算法框架,用于求解大规模线性规划(LP)问题。尽管基于PDHG的求解器在单节点GPU架构上表现出色,但其在工业规模实例中的适用性常受限于单GPU计算吞吐量。为克服这些挑战,我们提出了D-PDLP,首个分布式PDLP框架,通过约束矩阵的二维网格划分将PDHG扩展到多GPU环境。为提高负载平衡和计算效率,我们引入了块状随机排列策略结合非零元素意识矩阵划分。通过将PDHG迭代所需的密集计算分布化,所提出的框架利用多GPU并行性,以相对低的通信开销实现显著加速。在标准LP基准(包括MIPLIB和Mittelmann实例)以及大规模真实世界数据集上的广泛实验表明,基于cuPDLPx构建的分布式实现,在保持完整FP64数值精度的同时,实现了强扩展性和高性能。
英文摘要:
We present a distributed framework of the Primal-Dual Hybrid Gradient (PDHG) algorithm for solving massive-scale linear programming (LP) problems. Although PDHG-based solvers demonstrate strong performance on single-node GPU architectures, their applicability to industrial-scale instances is often limited by single-GPU computational throughput. To overcome these challenges, we propose D-PDLP, the first Distributed PDLP framework, which extends PDHG to a multi-GPU setting via a practical two-dimensional grid partitioning of the constraint matrix. To improve load balance and computational efficiency, we introduce a block-wise random permutation strategy combined with nonzero-aware matrix partitioning. By distributing the intensive computation required in PDHG iterations, the proposed framework harnesses multi-GPU parallelism to achieve substantial speedups with relatively low communication overhead. Extensive experiments on standard LP benchmarks (including MIPLIB and Mittelmann instances) as well as huge-scale real-world datasets show that our distributed implementation, built upon cuPDLPx, achieves strong scalability and high performance while preserving full FP64 numerical accuracy.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。