Re: Informix vs. Sybase vs. Oracle vs. (gasp) MS SQL Server

From: Pablo Sanchez <pablo_at_sgi.com>
Date: 1997/12/01
Message-ID: <yutafeka8zb.fsf_at_mew.corp.sgi.com>#1/1


>>>>> "David" == David Williams <djw_at_smooth1.demon.co.uk> writes:
David>
David> Psst! Wrong. Informix supports page and row level locking!

Well, yeah, but I was talking about page *only* versus having the ability to mix 'em. Sorry for being ambiguous.

David> If Informix can do either then performance comparision with David> Sybase mean nothing,

Not true. You see, during the bench, you bench for speed (among other things). So the bencher would set row level for those tables that would "need it"... therefore we have some tables using RLL compared to other tables using PLL.

David> if page was faster Informix would use David> page level locking in TPC tests!!

Right, sorta. I don't have the FDR for Informix on which tables were and were not using page level locks. What the bench points to is that with Sybase, doing PLL, it can (and does) perform as well (and this instance better) than an RDBMS that has the ability to do RLL.

David> Where is the comparision between Sybase and Informix on indentical David> hardware and dataset sizes? Let's get a like for like comparison.

Since a bench takes about two to four months to do, it'd be really tough to get that configuration you request. However you can use the numbers to show that RLL vs PLL isn't really an issue because if it were, then it wouldn't matter what size machine you throw at the problem... as I mentioned before.

-- 
Pablo Sanchez              | Ph # (650) 933.3812          Fax # (650) 933.2821
pablo_at_sgi.com              | Pg # (800) 930.5635   -or-   pablo_p_at_pager.sgi.com
-------------------------------------------------------------------------------
I am accountable for my actions.   http://reality.sgi.com/pablo [ /Sybase_FAQ ]
Received on Mon Dec 01 1997 - 00:00:00 CET

Original text of this message