A Study On The Wear Balance Algorithm Based On Divide Management Strategy

With the rapid development of modern information technology,storage in the computer field becomes more and more important-As a new storage medium,flash memory appeared.Its growing capacity,its tiny size,its high efficiency on reading and writing,its power saving and low power consumption,all these characters shall make the solid state drive(SSD)which is based on flash memory the next generation of mainstream storage devices,and then replace the traditional hard drive.But the life of flash memory is limited to the limited number of times of reading and writing,that will affect the reliability of solid-state hard drive.The rapid development of flash memory technology lead solid-state hard drive into the era of TB capacity,it's being more and more widely used.In order to improve the performance of solid-state hard drive and to extend its working life,wear balance algorithm was put forward.Aiming at the high cost of high-capacity NAND Flash-based SSD,which is effective in the process of garbage collection and wear balance,this paper proposes a wear balance algorithm based on divide and management strategy Wbabodms(wear balance algorithm based on Divide management strategy.As our demand for solid-state hard drives is increasing,its capacity is growing,and the overhead of increased capacity is growing.Taking into account the flash memory controller support copy back data migration mode in the process of data migration does not take up the characteristics of the bus,this paper uses copy_back rapid migration,put forward a divide management strategy,trying to solve the effective page data in garbage collection and wear The process of moving the cost of moving.This article uses the trace driver simulator SSDsim to verify the experiment.The experimental results show that the Wbabodms algorithm is reduced by 41.5%compared with the Dual Pool algorithm,which is 13.1%lower than the Rejuvenator algorithm for the average erasure times of the physical blocks for the Financial,Exchange,Cello,TraceA and TraceB loads.In the garbage collection and wear balance The Wbabodms algorithm is reduced by 81.6%over the Dual Pool algorithm and 25.7%less than the Rejuvenator algorithm.The Wbabodms algorithm is more efficient than the Dual Pool algorithm in the garbage collection and wear balance process.A decrease of 61.7%,a 14.3%reduction over the Rejuvenator algorithm,and a 23.1%increase in the effective data migration rate.