面向稀疏矩阵运算的位图掩码筛选缓存方法
By using a bitmap mask filtering caching method, the problems of wasted cache resources and low memory access bandwidth utilization in sparse matrix operations are solved, achieving efficient data filtering and storage optimization, and improving computing performance.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG UNIV
- Filing Date
- 2026-05-08
- Publication Date
- 2026-07-17
AI Technical Summary
Existing block caching mechanisms suffer from wasted cache resources and low memory access bandwidth utilization when handling sparse matrix operations, resulting in a large amount of redundant and invalid data occupying hardware storage resources and reducing computing performance.
A bitmap mask filtering and caching method is adopted. The bitmap mask is generated through preprocessing and then filtered and cached in real time by a three-level hardware pipeline at the hardware acceleration end to remove redundant data. The effective data is stored by using an address compression mapping strategy.
It improves cache payload utilization, enhances bus memory access bandwidth utilization, reduces on-chip storage footprint, does not introduce additional memory access latency, and optimizes computing performance.
Smart Images
Figure CN122152767B_ABST