1 of 1

MAPredict: Static Analysis Driven Memory Access Prediction Framework for Modern CPUs and GPUs

Scientific Achievement

Investigate the impact of different memory access patterns on the last-level cache (LLC)-memory traffic of different CPUs and GPUs and identify the similarities and dissimilarities in LLC-memory traffic on different generations of Intel CPUs, NVIDIA GPUs, and AMD GPUs.

Significance and Impact

Cache hierarchy is a significant role in deciding the compute and memory intensity of a program. Exploring when and why an LLC-memory transaction occurs is essential to understanding the performance on modern CPUs and GPUs. This study presents an approach to explore and understand the impact of memory-access patterns on various Intel CPUs, NVIDIA GPUs, and AMD GPUs.

Workflow of MAPredict Framework

Technical Approach

    • Find common factors in the cache hierarchy that trigger an LLC-memory transaction.
    • Investigate and compare four Intel CPUs (Broadwell, Skylake, Cascade Lake, Cooper Lake), three NVIDIA GPUs (P100, V100, A100), and three AMD GPUs (MI50, MI60, MI100), for different memory access patterns.
    • Develop a static analysis-driven framework named MAPredict to predict LLC-DRAM traffic at compile time.
    • Evaluate the LLC-memory traffic prediction accuracy of MAPredict for Intel CPUs and NVIDIA GPUs and AMD GPUs.

M. A. H. Monil, S. Lee, J. S. Vetter, and A. D. Malony, MAPredict: Static Analysis Driven Memory Access Prediction Framework for Modern CPUs, the ISC High Performance (ISC 2022), 2022.

M. A. H. Monil, S. Lee, J. S. Vetter, and A. D. Malony, Comparing LLC-memory Traffic between CPU and GPU Architectures, RSDHA: Redefining Scalability for Diversely Heterogeneous Architectures, in conjunction with SC21, 2021.