On Mon, Nov 11, 2013 at 4:02 PM, Jed Brown <[email protected]> wrote:
> Matthew Knepley <[email protected]> writes: > > I rewrote insertion for DMLabel to use the hash table, and convert to > flat > > arrays after insertion. Now its all linear: > > Nice. Did you check how many hash operations per second this works out > to? Is this still the leading setup cost? > I did not check, but it would be interesting to check 3D since that it 4 part keys. It by far the leading cost in mesh setup, but matrix preallocation is 3x more expensive right now. Matt -- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener
