> >I once implemented "closest seek" and "2way-elevators" for
> >the Linux IDE driver, and saw marked response improvements
> >under heavy load. In practice, starvation of the edges was
> >a non-issue, though I'm sure a pathological case could be
> >conceived that might be noticeably malaffected.
>
History. The most marked improvement in Disk/IO obtained in the
days of the DEC ST-506 interface was to simply mark a disk into
logical zones. DEC used 16, regardless of size.
Pending I/O requests were queued. There were only 16 "less/greater"
compares when traversing the list. All LBNs that fell within
each zone were completed. The next pass, the next zone, etc. This
would repeat forever. There was no expensive "sort".
This reduced head movement. The modified page-writer sorted
I/O requests to improve page-file performance, but this was
specific to memory-management, not your basic disk I/O.
Cheers,
Dick Johnson
***** FILE SYSTEM WAS MODIFIED *****
Penguin : Linux version 2.1.128 on an i686 machine (400.59 BogoMips).
Warning : It's hard to remain at the trailing edge of technology.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/