lkml.org 
[lkml]   [2002]   [Nov]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: rbtree scores (was Re: [patch] deadline-ioscheduler rb-tree sort)
    Jens Axboe wrote:
    > > If it was worth it (I suspect not), you can make a data structure
    > > which has O(1) amortised insertion time for a number of common cases,
    > > such as runs of ascending block numbers. Seems a likely pattern for a
    > > filesystem...
    >
    > Fibonacci heaps, for instance. I looked into that as well. However, it's
    > actually more important to have (if possible) O(1) extraction than
    > insert. Extraction is typically run from interrupt context, when a
    > driver wants to requeue more requests because it has completed one (or
    > some). That was a worry with the rbtree solution. The linked list may
    > have had sucky O(N) insert, but extraction was a nice O(1). So far I
    > haven't been able to notice any regression in this area, regardless.

    There's a skip list variant which offers O(1) extraction from the head
    of the list, and probabilistic O(log n) insertion, fwiw.

    -- Jamie
    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.kernel.org
    More majordomo info at http://vger.kernel.org/majordomo-info.html
    Please read the FAQ at http://www.tux.org/lkml/

    \
     
     \ /
      Last update: 2005-03-22 13:30    [W:0.032 / U:62.440 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site