Towards Performance-Aware Allocation for Accelerated Machine Learning on GPU-SSD Systems
Georgia Institute of Technology
arXiv:2412.04569 [cs.AR], (9 Dec 2024)
@misc{gundawar2024performanceawareallocationacceleratedmachine,
title={Towards Performance-Aware Allocation for Accelerated Machine Learning on GPU-SSD Systems},
author={Ayush Gundawar and Euijun Chung and Hyesoon Kim},
year={2024},
eprint={2412.04569},
archivePrefix={arXiv},
primaryClass={cs.AR},
url={https://arxiv.org/abs/2412.04569}
}
The exponential growth of data-intensive machine learning workloads has exposed significant limitations in conventional GPU-accelerated systems, especially when processing datasets exceeding GPU DRAM capacity. We propose MQMS, an augmented in-storage GPU architecture and simulator that is aware of internal SSD states and operations, enabling intelligent scheduling and address allocation to overcome performance bottlenecks caused by CPU-mediated data access patterns. MQMS introduces dynamic address allocation to maximize internal parallelism and fine-grained address mapping to efficiently handle small I/O requests without incurring read-modify-write overheads. Through extensive evaluations on workloads ranging from large language model inference to classical machine learning algorithms, MQMS demonstrates orders-of-magnitude improvements in I/O request throughput, device response time, and simulation end time compared to existing simulators.
December 15, 2024 by hgpu
