Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!seismo!mimsy!chris
From: chris@mimsy.UUCP
Newsgroups: comp.unix.questions
Subject: Re: Two questions about DBM
Message-ID: <7072@mimsy.UUCP>
Date: Tue, 16-Jun-87 11:19:54 EDT
Article-I.D.: mimsy.7072
Posted: Tue Jun 16 11:19:54 1987
Date-Received: Thu, 18-Jun-87 03:24:31 EDT
References: <662@vixie.UUCP>
Organization: U of Maryland, Dept. of Computer Science, Coll. Pk., MD 20742
Lines: 19

In article <662@vixie.UUCP> paul@vixie.UUCP (Paul Vixie Esq) writes:
>is NDBM faster than DBM,

No.

>or does it just do more things?

More databases, anyway.  It also has a nondestructive store operation
(avoids `fetch, make sure null, store').

>does [N]DBM do better or worse if adding lots of records in SORTED order?

It makes no noticeable difference what order things are stored.
What *would* help is some way of specifying in advance which blocks
will be split.  This is nontrivial, although simply telling dbm
the ultimate size of the database might suffice.
-- 
In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690)
Domain:	chris@mimsy.umd.edu	Path:	seismo!mimsy!chris