arXiv subjects

Huawen Liang

Publications and source records attributed to Huawen Liang.

3 recordsLinked to original sources

arXiv2025

Exploring the Performance Improvement of Tensor Processing Engines through Transformation in the Bit-weight Dimension of MACs

General matrix-matrix multiplication (GEMM) is a cornerstone of AI computations, making tensor processing engines (TPEs) increasingly critical in GPUs and domain-specific architectures. Existing architectures primarily optimize dataflow or operand reuse strategies. However, considering the interaction between matrix multiplication and multiply-accumulators (MACs) offers greater optimization potential. This work introduces a novel hardware perspective on matrix multiplication, focusing on the bit-weight dimension of MACs. We propose a finer-grained TPE notation using matrix triple loops as an example, introducing new methods for designing and optimizing PE microarchitectures. Based on this notation and its transformations, we propose four optimization techniques that improve timing, area, and power consumption. Implementing our design in RTL using the SMIC-28nm process, we evaluate its effectiveness across four classic TPE architectures: systolic array, 3D-Cube, multiplier-adder tree, and 2D-Matrix. Our techniques achieve area efficiency improvements of 1.27x, 1.28x, 1.56x, and 1.44x, and energy efficiency gains of 1.04x, 1.56x, 1.49x, and 1.20x, respectively. Applied to a bit-slice architecture, our approach achieves a 12.10x improvement in energy efficiency and 2.85x in area efficiency compared to Laconic. Our Verilog HDL code, along with timing, area, and power reports, is available at https://github.com/wqzustc/High-Performance-Tensor-Processing-Engines

cs.AR↗

arXiv2024

EN-T: Optimizing Tensor Computing Engines Performance via Encoder-Based Methodology

Tensor computations, with matrix multiplication being the primary operation, serve as the fundamental basis for data analysis, physics, machine learning, and deep learning. As the scale and complexity of data continue to grow rapidly, the demand for tensor computations has also increased significantly. To meet this demand, several research institutions have started developing dedicated hardware for tensor computations. To further improve the computational performance of tensor process units, we have reexamined the issue of computation reuse that was previously overlooked in existing architectures. As a result, we propose a novel EN-T architecture that can reduce chip area and power consumption. Furthermore, our method is compatible with existing tensor processing units. We evaluated our method on prevalent microarchitectures, the results demonstrate an average improvement in area efficiency of 8.7\%, 12.2\%, and 11.0\% for tensor computing units at computational scales of 256 GOPS, 1 TOPS, and 4 TOPS, respectively. Similarly, there were energy efficiency enhancements of 13.0\%, 17.5\%, and 15.5\%.

cs.AR↗

arXiv2024

Temporal Interaction and its Role in the Evolution of Cooperation

This research investigates the impact of dynamic, time-varying interactions on cooperative behaviour in social dilemmas. Traditional research has focused on deterministic rules governing pairwise interactions, yet the impact of interaction frequency and synchronization in groups on cooperation remains underexplored. Addressing this gap, our work introduces two temporal interaction mechanisms to model the stochastic or periodic participation of individuals in public goods games, acknowledging real-life variances due to exogenous temporal factors and geographical time differences. We consider that the interaction state significantly influences both game payoff calculations and the strategy updating process, offering new insights into the emergence and sustainability of cooperation. Our results indicate that maximum game participation frequency is suboptimal under a stochastic interaction mechanism. Instead, an intermediate activation probability maximizes cooperation, suggesting a vital balance between interaction frequency and inactivity security. Furthermore, local synchronization of interactions within specific areas is shown to be beneficial, as time differences hinder the spread of cross-structures but promote the formation of dense cooperative clusters with smoother boundaries. We also note that stronger clustering in networks, larger group sizes and lower noise increase cooperation. This research contributes to understanding the role of node-based temporality and probabilistic interactions in social dilemmas, offering insights into fostering cooperation.

physics.soc-ph↗