On 2/13/2013 1:37 PM, John Gilmore wrote:
When I first heard about the scheme Edward Jaffe has just mentioned I
did some simulations. If optimality is identified with extents-used
minimization they suggest that the new scheme will work better than
the old one but not nearly so well as a full optimization, which would
require the solution of an integer linear-programming problem having
smallish embedded knapsack problems, a variant of the newsprint-roll
cutting problem.
But would that not require "active" data movement/cleanup whenever a block is
freed?
I think they settled for something they could pull off without radical
re-architecture to the library or its serialization mechanisms.
--
Edward E Jaffe
Phoenix Software International, Inc
831 Parkview Drive North
El Segundo, CA 90245
http://www.phoenixsoftware.com/
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
