IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · 2023 · 14 citations · 30 references
EngineeringComputer EngineeringComputer ScienceComputer ArchitectureHardware-software Co-designMemoryData Movement OptimizationHardware AccelerationMemory HierarchyMemory Hierarchy DesignMemory Hierarchy OptimizationMemory Subsystem DesignData Flow ManagementGraph AggregationGraph ProcessingHardware-software Co-optimizationMemory Access PatternsNonvolatile MemoryTemporal LocalityVon Neumann Architecture
Aggregating features from neighbor vertices is a fundamental operation in graph convolution network (GCN). However, the sparsity in graph data creates poor spatial and temporal locality, causing dynamic and irregular memory access patterns and limiting the performance of aggregation on the Von Neumann architecture. The emerging processing-in-memory (PIM) architecture is based on emerging nonvolatile memory (NVM), like spin-orbit torque magnetic RAM (SOT-MRAM), and demonstrates promising prospects in alleviating the Von Neumann bottleneck. However, the limited memory capacity of PIM medium still incurs non-negligible data movements between PIM architecture and external memory. To solve this challenge, we propose an SOT-MRAM-based in-memory computing architecture, called IMGA, for efficient in-situ graph aggregation. Specifically, we design adaptive data flow management strategies that reuse vertex data in MRAM when processing graphs of different scales and adopt edge data as the control signal source to utilize the graph’s structural information. A reordering optimization strategy leveraging hardware–software co-design principle is proposed to further reduce the costly data movement. Experimental results demonstrate that IMGA achieves an average <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$2523\times $ </tex-math></inline-formula> and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$21\times $ </tex-math></inline-formula> speedup, and 1.03E+6 and 1.04E+3 energy efficiency compared with CPU and GPU, respectively.
30
Overcoming the challenges of crossbar resistive memory architectures
Cong Xu, Dimin Niu, Naveen Muralimanohar et al. · 2015 · 331 citations
HyGCN: A GCN Accelerator with Hybrid Architecture
Mingyu Yan, Lei Deng, Xing Hu et al. · 2020 · 331 citations
GraphR: Accelerating Graph Processing Using ReRAM
Linghao Song, Youwei Zhuo, Xuehai Qian et al. · 2018 · 294 citations
AWB-GCN: A Graph Convolutional Network Accelerator with Runtime Workload Rebalancing
Tong Geng, Ang Li, Runbin Shi et al. · 2020 · 285 citations