Key-value （KV） stores have become a backbone of large-scale applications in today＇s data centers. Write- optimized data structures like the Log-Structured Merge-tree （LSM-tree） and their variants are widely used ...Key-value （KV） stores have become a backbone of large-scale applications in today＇s data centers. Write- optimized data structures like the Log-Structured Merge-tree （LSM-tree） and their variants are widely used in KV storage systems like BigTable and RocksDB. Conventional LSM-tree organizes KV items into multiple, successively larger components, and uses compaction to push KV items from one smaller component to another adjacent larger component until the KV items reach the largest component. Unfortunately, current compaction scheme incurs significant write amplification due to repeated KV item reads and writes, and then results in poor throughput. We propose a new compaction scheme, delayed compaction （dCompaction） that decreases write amplification, dCompaction postpones some compactions and gathers them into the following compaction. In this way, it avoids KV item reads and writes during compaction, and consequently improves the throughput of LSM-tree based KV stores. We implement dCompaction on RocksDB, and conduct extensive experiments. Validation using YCSB framework shows that compared with RocksDB, dCompaction has about 40% write performance improvements and also comparable read performance.展开更多
大数据对存储系统的可扩展性、性能和成本等方面提出了更高的要求。瓦记录(Shingled Magnetic Recording,SMR)硬盘由于存储密度高、价格便宜,正逐步被广泛应用于大数据存储系统。但是,SMR硬盘的随机写性能较差,与快速的基于闪存的固态硬...大数据对存储系统的可扩展性、性能和成本等方面提出了更高的要求。瓦记录(Shingled Magnetic Recording,SMR)硬盘由于存储密度高、价格便宜,正逐步被广泛应用于大数据存储系统。但是,SMR硬盘的随机写性能较差,与快速的基于闪存的固态硬盘(Solid State Drive,SSD)一起构成混合存储时可以显著提升性能。同时,基于写优化的日志结构合并(Log-Structured Merge,LSM)树的键值存储已被广泛应用于许多NoSQL系统,如BigTable,Cassandra和HBase等。因此,如何基于新型的SSD-SMR混合存储构建出高性能的LSM树键值存储系统是一个具有很大研究价值的问题。首先建立基于SSD-SMR混合存储的LSM树键值系统的性能模型,然后针对SSD和SMR的硬件特征以及LSM树键值存储的软件特点,设计了一套面向SSD-SMR混合存储进行性能优化的LSM树键值存储系统,并基于LevelDB实现了该系统。在仅仅使用0.4%~2%空间的SSD的情况下,所提方法可以使SSD-SMR混合存储方案比普通磁盘方案的随机写性能提高20%,随机读性能提高5倍。展开更多
基金This work is supported by the National Key Research and Development Program of China under Grant No. 2016YFB1000202 and the National Natural Science Foundation of China under Grant Nos. 61303056 and 61379042.
文摘Key-value （KV） stores have become a backbone of large-scale applications in today＇s data centers. Write- optimized data structures like the Log-Structured Merge-tree （LSM-tree） and their variants are widely used in KV storage systems like BigTable and RocksDB. Conventional LSM-tree organizes KV items into multiple, successively larger components, and uses compaction to push KV items from one smaller component to another adjacent larger component until the KV items reach the largest component. Unfortunately, current compaction scheme incurs significant write amplification due to repeated KV item reads and writes, and then results in poor throughput. We propose a new compaction scheme, delayed compaction （dCompaction） that decreases write amplification, dCompaction postpones some compactions and gathers them into the following compaction. In this way, it avoids KV item reads and writes during compaction, and consequently improves the throughput of LSM-tree based KV stores. We implement dCompaction on RocksDB, and conduct extensive experiments. Validation using YCSB framework shows that compared with RocksDB, dCompaction has about 40% write performance improvements and also comparable read performance.
文摘大数据对存储系统的可扩展性、性能和成本等方面提出了更高的要求。瓦记录(Shingled Magnetic Recording,SMR)硬盘由于存储密度高、价格便宜,正逐步被广泛应用于大数据存储系统。但是,SMR硬盘的随机写性能较差,与快速的基于闪存的固态硬盘(Solid State Drive,SSD)一起构成混合存储时可以显著提升性能。同时,基于写优化的日志结构合并(Log-Structured Merge,LSM)树的键值存储已被广泛应用于许多NoSQL系统,如BigTable,Cassandra和HBase等。因此,如何基于新型的SSD-SMR混合存储构建出高性能的LSM树键值存储系统是一个具有很大研究价值的问题。首先建立基于SSD-SMR混合存储的LSM树键值系统的性能模型,然后针对SSD和SMR的硬件特征以及LSM树键值存储的软件特点,设计了一套面向SSD-SMR混合存储进行性能优化的LSM树键值存储系统,并基于LevelDB实现了该系统。在仅仅使用0.4%~2%空间的SSD的情况下,所提方法可以使SSD-SMR混合存储方案比普通磁盘方案的随机写性能提高20%,随机读性能提高5倍。
