Pathfinding Future PIM Architectures by Demystifying a Commercial PIM Technology

2023/08/01 by Bongjoon Hyun, Taehun Kim, Hyun, Bongjoon +5 · 11 citations
Computer Science · #Parallel Computing and Optimization Techniques #Advanced Data Storage Technologies #Distributed and Parallel Computing Systems

paper · pdf · doi:10.48550/arxiv.2308.00846

Abstract

Processing-in-memory (PIM) has been explored for decades by computer architects, yet it has never seen the light of day in real-world products due to their high design overheads and lack of a killer application. With the advent of critical memory-intensive workloads, several commercial PIM technologies have been introduced to the market ranging from domain-specific PIM architectures to more general-purpose PIM architectures. In this work, we deepdive into UPMEM's commercial PIM technology, a general-purpose PIM-enabled parallel architecture that is highly programmable. Our first key contribution is the development of a flexible simulation framework for PIM. The simulator we developed (aka PIMulator) enables the compilation of UPMEM-PIM source codes into its compiled machine-level instructions, which are subsequently consumed by our cycle-level performance simulator. Using PIMulator, we demystify UPMEM's PIM design through a detailed characterization study. Building on top of our characterization, we conduct a series of case studies to pathfind important architectural features that we deem will be critical for future PIM architectures to support

Cited by

• PIMID: A Full-System Simulator with Intricacy and Diversity for Processing-in-Memory
• DCC: Data-Centric Compilation of Machine Learning Kernels for Processing-In-Memory Architectures
• CompAir: Synergizing Complementary PIMs and In-Transit NoC Computation for Efficient LLM Acceleration
• IM-PIR: In-Memory Private Information Retrieval
• CoMoNM: A Cost Modeling Framework for Compute-Near-Memory Systems
• Pimba: A Processing-in-Memory Acceleration for Post-Transformer Large Language Model Serving
• Accelerating Triangle Counting with Real Processing-in-Memory Systems
• PIM-malloc: A Fast and Scalable Dynamic Memory Allocator for Processing-In-Memory (PIM) Architectures
• Sparse Attention Remapping with Clustering for Efficient LLM Decoding on PIM
• Memory-Centric Computing: Solving Computing's Memory Problem
• L3: DIMM-PIM Integrated Architecture and Coordination for Scalable Long-Context LLM Inference

Related

• PIMID: A Full-System Simulator with Intricacy and Diversity for Processing-in-Memory
• Modeling and Simulation Frameworks for Processing-in-Memory Architectures
• SimplePIM: A Software Framework for Productive and Efficient Processing-in-Memory
• New Tools, Programming Models, and System Support for Processing-in-Memory Architectures
• Methodologies, Workloads, and Tools for Processing-in-Memory: Enabling the Adoption of Data-Centric Architectures