Recently, the Network and Distributed Systems Laboratory, part of the School of Computing (specializing in Network Security), has achieved remarkable strides in high-performance computing and GPU architecture optimization. Its related accomplishments have been consistently published in CCF Category A conferences and journals within the computer architecture domain. The lab has focused its research on critical challenges, including high-performance matrix factorization algorithms tailored for GPU architectures, singular value decomposition of low-rank and rank-deficient matrices, as well as the design of one-sided linear algebra libraries. Three of its papers have been accepted by prestigious platforms: PPoPP 2026, IEEE Transactions on Parallel and Distributed Systems (TPDS), and ACM Transactions on Architecture and Code Optimization. These publications have significantly bolstered the lab's academic influence in high-performance computing, parallel algorithms, and GPU numerical software.