I think it means LRU assumes reference requests are going to happen in clusters ( groups). i.e probability of getting reference request is more if it was referenced in immediate past. with regards Prasad Hegde On 8/6/07, akshay sumant <sumantakshay@xxxxxxxxx> wrote: > While reading about "Adaptive replacement cache" i came across this line... > "LRU maintains cache pages based on their most recent access times and > replaces the least recently used pages. While LRU is simple to implement and > captures the "clustered locality of reference" that is common in many > workloads" > What is meaning of clustered locality of reference?? i know about spatial > and temporal locality of reference.... but didn't got this one.. > Pls explain.. Thank you -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
