Path: news.easynews.com!core-easynews!newsfeed1.easynews.com!easynews.com!easynews!news-feed01.roc.ny.frontiernet.net!nntp.frontiernet.net!logbridge.uoregon.edu!newsfeed.vmunix.org!newsfeed01.sul.t-online.de!newsmm00.sul.t-online.com!t-online.de!news.t-online.com!not-for-mail
From: "Carl Rosenberger" <carl@db4o.com>
Newsgroups: comp.databases.theory
Subject: Re: Alternatives to ACID, non-serialized concurrent transactions?
Date: Mon, 5 May 2003 19:59:12 +0200
Organization: T-Online
Lines: 39
Message-ID: <b96848$oh4$04$1@news.t-online.com>
References: <af3d9224.0305041920.e78b7e9@posting.google.com> <c0d87ec0.0305050824.12d3cbf3@posting.google.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: news.t-online.com 1052156873 04 25124 4iF7EkrGSGY2oV 030505 17:47:53
X-Complaints-To: abuse@t-online.com
X-ID: T80nc6Z18er8SqJu8D4Fi991EOT3EGSiT-JoJ6303P8XjYUq6vsXEp
X-Priority: 3
X-MSMail-Priority: Normal
X-Newsreader: Microsoft Outlook Express 6.00.2462.0000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2462.0000
Xref: core-easynews comp.databases.theory:26183
X-Received-Date: Mon, 05 May 2003 10:48:37 MST (news.easynews.com)


My compliments!
I really liked the original article, especially the consideration
that there may be micro- and macro- views on the data. It would
be possible to allow different concurrency strategies for different
types of clients:

(1) Client 1 could have a macro view, to do an overall calculation
over the number of flights booked in a certain time period. He would
not need to care about locks. What he would need would be mechanism
to read data as it existed to a certain point in time.

(2) Client 2 would need to make sure, a specific seat on the plane
is still available, so he would need a micro view.


Just one side note her:

"--CELKO--" wrote:
> There are scheduling problems that need to be solved so that every
> session gets to use the database, the overall work flow is optimal or
> near-optimal, etc.

Some object databases can solve this problem with object-level locking.

I am thinking about a "locking-by-example" concept:
You construct a graph of objects as an example to describe the depth
of objects and all the child objects that are to be locked. This
"lock all as the example" could be applied to a query result.


Kind regards,
Carl
--
Carl Rosenberger
db4o - database for objects - http://www.db4o.com



