FTL Flash Memory Data and Clusters

30 May

In this paper, we designed and implemented a flash translation layer and data management mechanisms, in flash memory embedded systems or mobile devices quick cleaning. We base ourselves on DAC technology and partitions hot flash is divided into three areas identified FTL flash memory data and separate clusters of hot and cold data. We then heat and cold are stored into different blocks of data in accordance with attribute data, to reduce system overhead, the cleaning activity.
At the same time, we use multiple block-sorting last a failure when, in the same area of the LRU list. In addition, we use cost effective policy, considering the weight of the block to select the most appropriate recovery block space.
Our proposed flash memory management schemes, quick cleaning mechanism, a series of experiments measured different workloads and a variety of flash memory utilization. Simulation results show that we propose fast cleansing policy incurs decrease of 0.6% -19.63% reduction in the erase operation, 1.7% -45.26% fewer pages copied, and 43.21% -57.43%, cleaning time than other high locality of reference according to policy DAC technology under flash and 90% utilization, we recommend flash quick cleaning mechanism management plan, incurring 0.6% -28.32% reduction erase operation, 1.7% -57.28% fewer pages copied, and 43.21% -65.17% reduce cleaning time than other flash management plan with different cleaning mechanisms under high locality reference, 90 percent of flash memory utilization.
In the future, we will measure our proposed flash memory management and quick cleaning, use a real NAND-type flash memory and test a variety of realistic workload performance improvements. The memory required to keep the information.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: