PIMSIM-NN: An ISA-based Simulation Framework for Processing-in-Memory Accelerators
Abstract: Processing-in-memory (PIM) has shown extraordinary potential in accelerating neural networks. To evaluate the performance of PIM accelerators, we present an ISA-based simulation framework including a dedicated ISA targeting neural networks running on PIM architectures, a compiler, and a cycleaccurate configurable simulator. Compared with prior works, this work decouples software algorithms and hardware architectures through the proposed ISA, providing a more convenient way to evaluate the effectiveness of software/hardware optimizations. The simulator adopts an event-driven simulation approach and has better support for hardware parallelism. The framework is open-sourced at https://github.com/wangxy-2000/pimsim-nn.
- Z. Zhu et al., “MNSIM 2.0: A Behavior-Level Modeling Tool for Memristor-based Neuromorphic Computing Systems,” in GLSVLSI, 2020.
- A. Ankit et al., “PUMA: A Programmable Ultra-efficient Memristor-based Accelerator for Machine Learning Inference,” in ASPLOS, 2019.
- X. Chen, “Instruction set architecture (isa) for processing-in-memory dnn accelerators,” arXiv preprint arXiv:2308.06449, 2023.
- X. Sun et al., “PIMCOMP: A Universal Compilation Framework for Crossbar-based PIM DNN Accelerators,” in DAC, 2023.
- S. K. Mandal et al., “A latency-optimized reconfigurable noc for in-memory acceleration of dnns,” JETCAS, vol. 10, no. 3, pp. 362–375, 2020.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.