Yuhong Dan

Yuhong Dan

Auto-generated profile, based on publicly available information

Yuhong Dan

School of Computer and Big Data Science, Jiujiang University, Jiujiang, China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 September 2025 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
CYchao yin
chao yin
ZXZhiyuan Xu
Zhiyuan Xu
YDYuhong Dan
Yuhong Dan
WQWanglong Qiu
Wanglong Qiu
ZHZhijie Huang
Zhijie Huang
WGWang Guangxing
Wang Guangxing
Distributed erasure coding Internet of Things (IoT) clusters can store data and achieve data reliability. This technology achieves high fault tolerance by storing data chunks and parity chunks. However, current storage systems often face the problem of unsatisfactory storage space and bandwidth utilization due to narrow stripes. To address this, we propose the stripe-merging algorithm NStripeMerge, which consists of two optimization algorithms: NStripeMerge-C and NStripeMerge-LB. These two algorithms address the minimum merging costs and the load balance problems during wide-stripe generation. We demonstrate that both algorithms can achieve minimum merging costs overhead and the highest node utilization under different conditions. NStripeMerge-C can reduce bandwidth without degrading performance. NStripeMerge-LB, on the other hand, achieves better load balance. Our experiments show that NStripeMerge-C can reduce merging costs by up to 12.4%, while NStripeMerge-LB can increase node utilization by up to 24.2% over a state-of-the-art storage scaling approach. Our proposed solution provides a promising approach for improving stripe-merging performance in erasure-coded storage systems.