2022上半年论文计划
2022上半年计划 (2月~10月)
- 重点论文:面向强化学习/大模型的存储优化
- 自适应索引:AegisKV: A High-Performance Persistent Key-Value Store via Adaptive Generated Index
DAC 2月第二周出结果中(0.1%几率):欢天喜地不中:修改再投ICPP
- ICPP投稿
ADSTS: Automatic Distributed Storage Tuning System Using Deep Reinforcement Learning6月12日一审
- IPDPS论文转投期刊
- Learning Data Placement with RLDP in Modern Distributed Storage Systems
- 存储和AI综述
- Storage Meets Artificial Intelligence: A Survey
- TridentKV的Scan优化
- AegisKV: A Range-query Optimized Key-Value Store via Efficient Partitioning and Learned Range Filter
计划会议:ICPP(4月),Cluster (5月),IPDPS(10月)等等
数据分布方案投期刊LibraDP: Toward Balanced and Controlled Data Placement in Large-scale Object Storage Systems.
Scan Filter论文计划期刊:TPDS计划会议:Neurips(5月)
毕业论文
分布式存储系统智能化关键技术研究智能化分布式存储系统关键技术研究基于AI的分布式存储优化的关键技术研究- 基于机器学习的分布式存储系统优化研究
针对存储系统的性能/可靠性/可用性
- 基于强化学习的分布式存储数据分布优化
- 基于自适应学习索引的键值存储读性能优化
- 一种高效的分布式存储自动调参框架研究和实现
万继光,王中华,鲁凯,李大平,郑文凯,谭志虎,谢长生,一种基于循环神经网络的磁盘故障预测方法,2020-5-13,中国,专利号: ZL2018113984454
This work was sponsored in part by the Key Research and Development Program of Guangdong Province under grant No. 2021B0101400003, the National Natural Science Foundation of China under Grant No.62072196, the Creative Research Group Project of NSFC No.61821003.