| Memcached is a high-performance, distributed memory object caching system. Itcaches objects in memory to reduce the number of database reads and improve theread andwrite speeds. With the rapid development of web applications, Memcached is requiredofmore efficiency and concurrency performance and is playing an increasingly importantrole in the field of cache.Then an urgent problem occursof seeking an optimizationmechanismto meet the needs of application services and performance.Theresult shows thedifficultiesof balancing memory efficiency and high concurrentrequests using small object key storage and read-oriented workloads by analyzingthecharacteristic ofMemcached workloads and the comparisonbetween different systems inmemory efficiency and concurrency mechanism. This papperprovides thecorrespondingsolution named optimization of cuckoo hashing algorithm to replacethedefault system hash table. It can enhance the efficiency of Memcached memorywithhigh occupancy hash table, achieve a single writer multiple readers concurrent access tohash table with optimized hash table queryinsertion algorithm and maintain the operationof hash table cache friendliness with tagbasedquery and insertion. The optimisticconcurrency controlleruses counter to replace the original system global lock and useschangeof version number to find concurrent replacement operation, then it canreduceinternal thread synchronization, ensure data consistency,support a large number ofconcurrent access and reduce insert query overhead.We achieved a memory efficiency and concurrency performance optimization inMemcached basing on cup-intensive computing with linux operating platform.Theoptimization improves multi readers single writer high concurrent access rateswhilemaintaining a high efficiency of the memory, decreases the memory usage by20%withsame test data setwhich meets zipfan distribution and increases the throughput by at leasttwo-foldwith high concurrency of client requests in reading and writing. |
PDF Full Text Request