ATA-Cache: Contention Mitigation for GPU Shared L1 Cache With Aggregated Tag Array
- 29 November 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 43 (5) , 1429-1441
Abstract
To fully exploit the locality of GPU applications, the GPU shared L1 cache architecture, which shares L1 cache among multiple GPU cores, is a promising architecture while still suffering from high-resource contentions. We present a GPU shared L1 cache architecture with an aggregated tag array that minimizes the L1 cache contentions and takes full advantage of inter-core locality. The key idea is to decouple and aggregate the tag arrays of multiple L1 caches so that the cache requests can be compared with all tag arrays in parallel to probe the replicated data in other caches. The GPU caches are only accessed by other GPU cores when replicated data exists, filtering out unnecessary cache accesses that cause high-resource contentions. We also develop a two-level thread-block scheduling policy adapted for the shared L1 cache architecture to maximize the available locality. The experimental results show that GPU performance can be improved by 14.5% on average for applications with a high inter-core locality.Keywords
All Related Versions
Funding Information
- National Key Research and Development Program of China (2023YFB4503101)
- National Natural Science Foundation of China (62272026, 62104014, 62072432)
- State Key Laboratory of Software Development Environment (SKLSDE-2022ZX-07)
- Beihang Frontier Interdisciplinary Fund (YWF-23-Q-1015)
- Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing; and
- Iluvatar CoreX Semiconductor Company Ltd
This publication has 36 references indexed in Scilit:
- Analyzing and Leveraging Decoupled L1 Caches in GPUs2021
- Tensor Casting: Co-Designing Algorithm-Architecture for Personalized Recommendation Training2021
- Accel-Sim: An Extensible Simulation Framework for Validated GPU Modeling2020
- The Locality Descriptor: A Holistic Cross-Layer Abstraction to Express Data Locality In GPUs2018
- SIPT: Speculatively Indexed, Physically Tagged Caches2018
- Power Efficient Sharing-Aware GPU Data Management2017
- GpuTejas: A parallel simulator for GPU architectures2014
- GPU accelerated sparse matrix‐vector multiplication and sparse matrix‐transpose vector multiplicationConcurrency and Computation: Practice and Experience, 2014
- Improving GPGPU resource utilization through alternative thread block scheduling2014
- Rodinia: A benchmark suite for heterogeneous computing2009