Font Size: a A A

Research On Mass Data Storage Model Based On Heat Perception

Posted on:2018-11-05Degree:MasterType:Thesis
Country:ChinaCandidate:W K WangFull Text:PDF
GTID:2348330569986435Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
The era of big data is here,the demand of mass data for the storage and processing ability of computer system is bigger and bigger.The technologies of parallel computing,cloud computing and so on are maturing;and the performance of processor is significantly improved,so that the computer's information process ability is strong enough.However,the performance of computer storage system has not improved much.For example,traditional DRAM main memory integration is low.The performance gap between memory and secondary storage.For instance,because of the high density,non-volatile features of PCM brings technological change to computer storage system and it breaks the existing storage architecture hopefully.Due to the asymmetry of PCM read and limited life write,it cannot replace the traditional DRAM main memory.Therefore,building a hybrid storage system with DRAM becomes mainstream.Because the performance gap between the main memory and the secondary storage is inevitable.Therefore,mixing the SSD and HDD to be secondary storage becomes an effective solution to the problem.In this paper,we use the DRAM and PCM to build mixed main memory and use the SSD and HDD to build secondary storage to build a hybrid storage system.Aiming at the hit rate in hybrid main memory and the writing life of PCM,we research the page management algorithm;and aiming at the I/O delay in hybrid secondary storage,we research the page migration strategy.The main contents include the following points:1.The page partition management method based on thermal perception.We research the hybrid memory architecture based on PCM&DRAM,we propose a page partition management method based on heat perception.We use the operating mechanism that is similar with traditional CLOCK algorithm to ensure the system hit rate.And we lead into the recently twice concept of writing distance and combine with the page history information to accurately judge the hot or cold of pages.Then design the page migration management mechanism.By writing CLOCK linked list to track the page writing heat dynamic,move the hot page to DRAM.And reduce the number of PCM write to improve the life of PCM.Finally,it is verified by the simulation experiments that this method reduces the number of write times on PCM by 9.5%,while ensuring the hit rate.2.Migration strategy based on I/O computing.We research the secondary storage based on hybrid storage architecture.On the basis of the existing algorithms,a migration strategy based on I/O is proposed.And we lead into the traditional hot and cold concept to divide for data heat.On this basis,we add a transient state so that the page needs to be operated two times to change it from a hot page to a cold page.At the same time we lead into the concept of page orientation.By statistics I/O cost of page storage in different media to determine the page storage location.By building the GEM5+NVMain+Disksim+Flashsim simulation platform,we can evaluate system performance.The experimental results show that this method can effectively reduce the I/O delay and reduce the running time of the system by about 12.8%.
Keywords/Search Tags:Mass Data, Hybrid Storage, Heat Perception, Migration strategy
PDF Full Text Request
Related items