The LazyFTL Scalability

15 Apr

LazyFTL scalability
Flash memory is organized in a larger array, the designer may wish to control the size of SRAM to reduce costs. Based on this assumption, the performance when LazyFTL we further study on the size of the SRAM is reduced.
With all other variables held constant, we first reduce the size of the LRU cache. This will increase the GMT access overhead, because GMT cache only a small portion in the SRAM, only a small portion can be read or modified.
GMT-access overhead can be quantified cache
Miss ratio. Cache hit ratio first very slowly decreased until less than 5 cached in SRAM. This trend means that the LRU cache can be reduced to save money, or for urban rail transit, to save space.
Then, we have reduced the the enlightened and CBA size limit. Smaller open and the meaning of the CBA, the conversion operation to be performed more frequently, and may have some blocks before conversion, the conversion costs have been reduced. Similarly, other configuration is the same as before. As shown in Figure 6 (c), this threshold amount is halved, the maximum conversion costs increased slowly, unless the threshold is at a certain level (about 32). Therefore, if it is used in a real-time system LazyFTL strictly with a single request to the waiting time, the total size of the UBA and CBA should be maintained at a certain level or more.


Leave a Reply

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

You are commenting using your 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: