MARS: Memory Aware Reordered Source
Ishwar Bhati, Udit Dhawan, Jayesh Gaur, Sreenivas Subramoney, and Hong, Wang

TL;DR
MARS is a memory reordering architecture that enhances memory efficiency and bandwidth utilization in GPUs by reordering data streams based on row-buffer addresses, addressing interleaving issues.
Contribution
MARS introduces a lookahead-based reordering method to improve memory locality and bandwidth in GPU workloads without requiring detailed memory configuration knowledge.
Findings
Achieves 11% increase in memory bandwidth on microbenchmarks.
Effectively restores locality lost due to interleaved data streams.
Operates without specific knowledge of memory system details.
Abstract
Memory bandwidth is critical in today's high performance computing systems. The bandwidth is particularly paramount for GPU workloads such as 3D Gaming, Imaging and Perceptual Computing, GPGPU due to their data-intensive nature. As the number of threads and data streams in the GPUs increases with each generation, along with a high available memory bandwidth, memory efficiency is also crucial in order to achieve desired performance. In presence of multiple concurrent data streams, the inherent locality in a single data stream is often lost as these streams are interleaved while moving through multiple levels of memory system. In DRAM based main memory, the poor request locality reduces row-buffer reuse resulting in underutilized and inefficient memory bandwidth. In this paper we propose Memory-Aware Reordered Source (\textit{MARS}) architecture to address memory inefficiency arising…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsParallel Computing and Optimization Techniques · Advanced Data Storage Technologies · Cloud Computing and Resource Management
