Benchmarking a New Paradigm: An Experimental Analysis of a Real Processing-in-Memory Architecture

TL;DR

This paper provides the first comprehensive experimental analysis of a real-world processing-in-memory (PIM) architecture, evaluating its performance and capabilities using microbenchmarks and a diverse benchmark suite.

Contribution

It introduces the first detailed empirical characterization of a real PIM system and presents PrIM, a benchmark suite for diverse workloads.

Findings

Identified architecture limits such as compute throughput and memory bandwidth.

Provided insights into the performance of real-world PIM hardware.

Demonstrated the versatility of PIM across various application domains.

Abstract

Many modern workloads, such as neural networks, databases, and graph processing, are fundamentally memory-bound. For such workloads, the data movement between main memory and CPU cores imposes a significant overhead in terms of both latency and energy. A major reason is that this communication happens through a narrow bus with high latency and limited bandwidth, and the low data reuse in memory-bound workloads is insufficient to amortize the cost of main memory access. Fundamentally addressing this data movement bottleneck requires a paradigm where the memory system assumes an active role in computing by integrating processing capabilities. This paradigm is known as processing-in-memory (PIM). Recent research explores different forms of PIM architectures, motivated by the emergence of new 3D-stacked memory technologies that integrate memory with a logic layer where processing elements can be easily placed. Past works evaluate these architectures in simulation or, at best, with simplified hardware prototypes. In contrast, the UPMEM company has designed and manufactured the first publicly-available real-world PIM architecture. This paper provides the first comprehensive analysis of the first publicly-available real-world PIM architecture. We make two key contributions. First, we conduct an experimental characterization of the UPMEM-based PIM system using microbenchmarks to assess various architecture limits such as compute throughput and memory bandwidth, yielding new insights. Second, we present PrIM, a benchmark suite of 16 workloads from different application domains (e.g., linear algebra, databases, graph processing, neural networks, bioinformatics).