Subject: Re: [boost] Interest in B-tree library for Boost?
From: Beman Dawes (bdawes_at_[hidden])
Date: 2010-09-15 14:16:06
On Wed, Sep 15, 2010 at 12:15 PM, Giorgio Zoppi <giorgio.zoppi_at_[hidden]> wrote:
> So you plan to code B-Tree, B*Tree, B+tree.
I use the term "B-tree" as the generic name, as is the usual practice
in the literature.
The particular implementation is a B+ tree with leaf pages in a doubly
linked list. But I consider that an implementation detail subject to
change. Single link lists have concurrency advantages. My C
implementation used no linked lists at all to advantage.
A pure B-tree has advantages for sets, although perhaps not for multisets.
I'm not much interested in B*trees, for the usual reasons.
Boost list run by bdawes at acm.org, david.abrahams at rcn.com, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk