Journal of Computers · 2019 · 27 citations · 15 references
EngineeringComputer EngineeringComputer ScienceComputer ArchitectureDatabasesParallel ComputingCachingHash TablesInstruction-level ParallelismMemory HierarchyMemory Hierarchy DesignMemory Subsystem DesignCollision HandlingInstruction Set DesignKey-value StorageKey-value StoresMemory UtilizationSearch EfficiencySystem LatencySystem Latency Reduction
This paper proposes a high-performance memory key-value database Redis++. In the memory management mechanism, Redis++ can apply and release a fixed-size memory segment from the system. The data in each memory segment is stored consecutively, and the memory is reclaimed based on the profit evaluation value. Secondly, a cache-friendly hash index structure is designed and the structure uses two-level index which can solve the hash collision to complete per search which needs cache mapping only once if possible. In addition, using the SIMD instruction set to realize instruction-level parallelism, which speeds up the search efficiency of the secondary index. The experiments prove the effect of Redis++ on memory utilization, system latency, and throughput.
15
Benchmarking cloud serving systems with YCSB
Brian F. Cooper, Adam Silberstein, Erwin Tam et al. · 2010 · 3.6K citations
Workload analysis of a large-scale key-value store
Berk Atikoglu, Yuehai Xu, Eitan Frachtenberg et al. · 2012 · 904 citations
Distributed caching with memcached
Brad Fitzpatrick · Linux journal · 2004 · 513 citations
Cache craftiness for fast multicore key-value storage
Yandong Mao, Eddie Kohler, Robert Tappan Morris · 2012 · 465 citations · Full text