11 Sep
2010
11 Sep
'10
3:14 p.m.
On August 29, 2010 09:15:45 Don Stewart wrote:
http://hackage.haskell.org/trac/ghc/ticket/4277
Proposal: Significant performance improvements for Data.Map
Description
Milan Straka's recent [52] Haskell Symposium paper (PDF) shed light on the containers:Data.Map library, indicating there were both algorithmic and stylistic performance improvements to be made.
What about the prior discussions on this list regarding the balance invariant not necessary being maintained during deletion? There seemed to be a feeling that it would still work with a smaller balance factor. Someone had also said they were going to work out a complete proof and post it somewhere. I don't believe there has been any more followup though. Cheers! -Tyson