Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Avoiding any locks in SQL Servers - read and understand....its magic.

Re: Avoiding any locks in SQL Servers - read and understand....its magic.

From: Billy Verreynne <vslabs_at_onwe.co.za>
Date: 19 Aug 2003 04:03:23 -0700
Message-ID: <1a75df45.0308190303.1bb3b863@posting.google.com>


Guido Stepken <stepken_at_little-idiot.de> wrote  

> Locking is not neccessary, database performance increases very much,
> because clients (and server) do not have to wait for any lock to be
> released.

I'm still not convinced. Nor have you bothered to address my concerns. Or any of the other posters'.

Simplest of examples. A file with a single record. 10 processes reading and writing that _same_ record. Without locking, how do you guarantee that one process i/o does not trashes another process's i/o and that data integrity remains?

By introducing more overheards - increasing space/memory requirements to have multiple copies? By making the clients "more intelligent"? Adding aditional processes to manage the conflict and ensure that only a single correctly constructed copy of that record resides on disk?

And how does this address the issue where serialisation of process i/o is mandatory? Please illustrate in a no-nonsense bullshit-free fashion how this non-locking method solves this.

And again - show me the HOW/WHERE/WHEN/WHY of "Locks are evil" and how this convoluted method can do it better than just a plain normal lock, the way Oracle handles it?

Is it not interesting that it is exactly those vendor products you have named, that are those vendors who cannot and never had been able to get locking right... and had to resort to dirty reads, page locking and crap like that?

> Say goodbye to locking, its not elementary for any business process.
> Look at traffic lights. All people think, they are neccessary. They're
> not. Substitute them by roundabouts and you will see, there will be much
> less traffic delays. Only the producers of traffic lights profit. By the
> way, this fact was proven in mathematical institute of cologne, germany
> by a former student, who worked on traffic simulations, never became
> really popular, Siemens was not delighted as producer from traffic
> lights technology.

Well, let me throw this analogy back at you. I have seen many traffic circles (aka round-abouts) USING traffic lights!!!!, simply because they do not work as envisioned.. because the traffic flow is nowhere near as the ideal solution that was modeled.. because the drivers does not behave as they thought the ideal (fantasy) driver should behave...

To quote Morphues.. Welcome to the real world. But then you had to go and swallow the red pill.

--
Billy
Received on Tue Aug 19 2003 - 06:03:23 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US