Hello Everybody,
I recently a paper on Buffer cache management using temporal and spatial locality.
The main motive here is to make the disk accesses more sequential. So the pages that are sequential are evicted first and the pages that reside randomly on the disk are given previlege to remain in cache for longer time.
In the context here 'spatial locality' refers to the sequentiality of pages on the disk
and 'temporal locality' refers to the times the page has been recently accessed..
In the paper those people say that
Currently only temporal locality(number of times page is accessed) is used in maintaining the page cache.
But we can also use the spatial locality(here: sequentiality of pages on the disk).
Sequences of pages are formed in the inactive LRU list and the sequence with more sequential pages is evicted first.
This helps to make disk I/O more sequential bcoz random blocks get privilege to be in cache more so that if they are referenced again then disk I/o is not required to bring them again in the cache..bcoz it takes more time to get random pages from disk than the sequential pages.
The sequences that are made in the inactive list are sorted on the basis of their temporal(recency) and spatial(sequence) locality.
To check if the page is accessed in the recent time we can have a timestamp..
But when the pages are in the LRU lists(inactive list) how can we check the sequentiality of the pages?
-Adheer
