Read and Write Performance Optimization of Storage System Based on LSM-tree Key Value
In a write-intensive work environment,log-structured-merge(LSM-Tree)has gradually become a mainstream storage system,there are problems in LSM-tree such as slow read operation speed,high cost of write operation,and low efficiency of range query operation,etc.In view of these problems,this paper presents a study to improve the performance of LSM-tree,and optimize the read and write performance of key-value storage system based on LSM-tree,and proposes a read and write performance optimiza-tion strategy for LSM-tree-based key-value storage system,designs the vTree structure through the key-value separation strategy,and presents the combination of layer merging and negative inter layer merging,as well as the strategy of range query-optimized merging,so as to optimize the range query performance of the system,and adopt different compression structures in the LSM-tree and vTree to improve the system's read and write performance;the experimental results show that the read performance is improved by 30％compared with the RocksDB system,and the range query performance is improved by 10％compared with the RocksDB-vTree system.
NETL
NSTL
万方数据
