Scalability Limitations of Processing-in-Memory using Real System Evaluations
Open Access
- 11 June 2024
- journal article
- research article
- Published by Association for Computing Machinery (ACM) in ACM SIGMETRICS Performance Evaluation Review
- Vol. 52 (1) , 63-64
Abstract
Processing-in-memory (PIM) has been widely explored in academia and industry to accelerate numerous workloads. By reducing the data movement and increasing parallelism, PIM offers great performance and energy efficiency. A large amount of cores or nodes present in PIM provide massive parallelism and compute throughput; however, this also proposes challenges and limitations for some workloads. In this work, we provide an extensive evaluation and analysis of a real PIM system from UPMEM. We specifically target emerging workloads featuring collective communication, demonstrating its role as the primary limitation within current PIM architecture.Keywords
This publication has 5 references indexed in Scilit:
- Memory-Centric Computing with SK Hynix's Domain-Specific Memory2023
- 25.4 A 20nm 6GB Function-In-Memory DRAM, Based on HBM2 with a 1.2TFLOPS Programmable Computing Unit Using Bank-Level Parallelism, for Machine Learning Applications2021
- The true Processing In Memory accelerator2019
- Deep Learning Recommendation Model for Personalization and Recommendation SystemsarXiv, 2019
- Homomorphic Encryption for Arithmetic of Approximate Numbers2017