SC2021Top-tier venue
CAKE: matrix multiplication using constant-bandwidth blocks
Abstract
We offer a novel approach to matrix-matrix multiplication computation on computing platforms with memory hierarchies. Constantbandwidth (CB) blocks improve computation throughput for architectures limited by external memory bandwidth. Configuring the shape and size of CB blocks operating from within any memory hierarchy level (e.g., internal SRAM), we achieve high throughput while holding external bandwidth (e.g., with DRAM) constant. We explain how, surprisingly, CB blocks can maintain constant external bandwidth as computation throughput increases. Analogous to partitioning a cake into pieces, we dub our CB-partitioned system CAKE.
We show CAKE outperforms state-of-the-art libraries in computation time on real-world systems where external bandwidth represents a bottleneck, demonstrating CAKE's ability to address the memory wall. CAKE achieves superior performance by directly using theoretically optimal CB-partitioned blocks in tiling and scheduling, obviating the need for extensive design search.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
3 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
Cited by top-tier papers3
- Register Tiling for Unstructured Sparsity in Neural Network InferenceLucas Wilkinson, Kazem Cheshmi, Maryam Mehri DehnaviPLDI 2023 · 17 citations
- autoGEMM: Pushing the Limits of Irregular Matrix Multiplication on Arm ArchitecturesDu Wu, Jintao Meng, Wenxi Zhu, Minwen Deng et al.SC 2024 · 14 citations
- KAMI: Communication-Avoiding General Matrix Multiplication within a Single GPUHemeng Wang, Yang Du, Sidu Li, Xiaowen Tian et al.SC 2025 · 4 citations
Builds on2
- MCUNet: Tiny Deep Learning on IoT DevicesJi Lin, Wei-Ming Chen, Yujun Lin, John Cohn et al.NeurIPS 2020 · 827 citations
- Planaria: Dynamic Architecture Fission for Spatial Multi-Tenant Acceleration of Deep Neural NetworksSoroush Ghodrati, Byung Hoon Ahn, Joon Kyung Kim, Sean Kinzer et al.MICRO 2020 · 120 citations
Related papers
- Analytical characterization and design space exploration for optimization of CNNsRui Li, Yufan Xu, Aravind Sukumaran-Rajam, Atanas Rountev et al.ASPLOS 2021 · 52 citations
- UniNDP: A Unified Compilation and Simulation Tool for Near DRAM Processing ArchitecturesTongxin Xie, Zhenhua Zhu, Bing Li, Yukai He et al.HPCA 2025 · 9 citations
- NDPBridge: Enabling Cross-Bank Coordination in Near-DRAM-Bank Processing ArchitecturesBoyu Tian, Yiwei Li, Li Jiang, Shuangyu Cai et al.ISCA 2024 · 27 citations
- WET: Write Efficient Loop Tiling for Non-Volatile Main MemoryMohammad A. Alshboul, James Tuck, Yan SolihinDAC 2020 · 3 citations
- TAPMM: A Traffic-Aware Page Mapping Method for Multi-level NUMA SystemsFengkun Dong, Guoqing Xiao, Haotian Wang, Yikun Hu et al.DAC 2024
