Multi-threaded read ahead prediction by pattern recognition

Electrical computers and digital processing systems: memory – Storage accessing and control – Hierarchical memories

Patent

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

711213, 711217, G06F 1208

Patent

active

060702303

ABSTRACT:
The inventive prediction mechanism constructs a time ordered state space of the file accesses, and then searches the state space looking for file accesses that form either sequential or stride patterns. A cache is maintained to store information about existing patterns. The mechanism will review the cache to determine if a new access request is a continuation of an existing pattern before searching the entire state space for a new pattern. If the request is either a continuation of an existing pattern or forms a new pattern, then a read ahead is issued for the next block in the pattern. The cache and the state space are updated as each request arrives. The mechanism uses a trigger array to prevent the re-issuing of predictions. The mechanism also has a strength feature which is an indication how many times the pattern has been continued, and can be used by the operating system in deciding how many read aheads should be issued.

REFERENCES:
patent: 5287487 (1994-02-01), Priem et al.
patent: 5357618 (1994-10-01), Mirza et al.
patent: 5568651 (1996-10-01), Medina et al.
patent: 5694568 (1997-12-01), Harrison, III et al.
patent: 5822790 (1998-10-01), Mehrotra

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Multi-threaded read ahead prediction by pattern recognition does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Multi-threaded read ahead prediction by pattern recognition, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multi-threaded read ahead prediction by pattern recognition will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-1918933

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.