LSAF: A load-balancing SpGEMM acceleration framework with dynamic package and static partition for multi-core systolic arrays

Key Points

• Enhances performance in load-balancing matrix multiplications using spGEMM techniques and systolic arrays.
• Achieves reduced computation times and energy consumption through dynamic package and static partition strategies.
• Supports efficient allocation of computational resources by leveraging parallelism across multi-core architectures.
• Demonstrates effectiveness through simulation results, indicating significant improvements over previous methods.