ReFloat：用于加速迭代线性求解器的ReRAM低成本浮点处理
ReFloat: Low-Cost Floating-Point Processing in ReRAM for Accelerating Iterative Linear Solvers
AI总结：
ReFloat提出一种数据格式与加速器架构，在ReRAM中实现低成本浮点处理，加速迭代线性求解器，相比现有方案提速5.02至84.28倍。
摘要：
电阻式随机存取存储器（ReRAM）是一种有前景的技术，可以在模拟域中执行低成本的矩阵向量乘法（MVM）。科学计算需要高精度的浮点（FP）处理。然而，由于浮点数值范围大，在ReRAM中执行浮点计算因硬件成本高和执行时间长而具有挑战性。在这项工作中，我们提出了ReFloat，一种数据格式和加速器架构，用于在ReRAM中为迭代线性求解器实现低成本和高性能的浮点处理。ReFloat匹配ReRAM交叉阵列硬件，并以减少的比特数和优化的指数基数表示一组浮点值，以实现高动态表示范围。因此，ReFloat实现了更少的ReRAM交叉阵列消耗和更少的处理周期，并克服了先前工作中的不收敛问题。在SuiteSparse矩阵上的评估表明，与最先进的基于ReRAM的加速器相比，ReFloat在求解器时间上实现了5.02倍到84.28倍的改进。
英文摘要：
Resistive random access memory (ReRAM) is a promising technology that can perform low-cost and in-situ matrix-vector multiplication (MVM) in analog domain. Scientific computing requires high-precision floating-point (FP) processing. However, performing floating-point computation in ReRAM is challenging because of high hardware cost and execution time due to the large FP value range. In this work we present ReFloat, a data format and an accelerator architecture, for low-cost and high-performance floating-point processing in ReRAM for iterative linear solvers. ReFloat matches the ReRAM crossbar hardware and represents a block of FP values with reduced bits and an optimized exponent base for a high range of dynamic representation. Thus, ReFloat achieves less ReRAM crossbar consumption and fewer processing cycles and overcomes the noncovergence issue in a prior work. The evaluation on the SuiteSparse matrices shows ReFloat achieves 5.02x to 84.28x improvement in terms of solver time compared to a state-of-the-art ReRAM based accelerator.
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。
