.png){kind=icon}
Return
Designing Programmable Accelerators for Sparse Tensor Algebra
.png){kind=icon}
PRE
AI
DOI：10.1109/MM.2025.3556611.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Recent research has focused on leveraging sparsity in hardware accelerators to improve the efficiency of applications spanning scientific computing to machine learning. Most such prior accelerators are fixed-function, which is insufficient for two reasons. First, applications typically include both dense and sparse components, and second, the algorithms that comprise these applications are constantly evolving. To address these challenges, we designed a programmable accelerator called Onyx for both sparse tensor algebra and dense workloads. Onyx extends a coarse-grained reconfigurable array (CGRA) optimized for dense applications with composable hardware primitives to support arbitrary sparse tensor algebra kernels. In this article, we show that we can further optimize Onyx by adding a small set of hardware features for parallelization that significantly increase both temporal and spatial utilization of the CGRA, reducing runtime by up to 6.2×.
Keywords:
Tensors
Pipeline processing
Micromechanical devices
Repeaters
Random access memory
Arrays
Sparse matrices
Process control
Hardware acceleration
Product design
Journal
IF:
2.9
Papers:
141
Citations:
2.7K
Organization
Cited Papers
ShareSaveHardware Acceleration of Sparse and Irregular Tensor Computations of ML Models: A Survey and Insights.png)
PROCEEDINGS OF THE IEEE
OAAI
PROCEEDINGS OF THE IEEEIF25.9
OAAIShareSaveShareSaveShareSaveShareSaveShareSaveShareSaveShareSave
View more