Re: Database or store to handle 30 Mb/sec and 40,000 inserts/sec

Here is just one from:
http://www.orafaq.com/faq/RDBMS_Server/Internals/faq503.htm

1, "ITL shortage deadlocks"
2, High transaction activity on tables with bitmap indexes.

Would you like me find and quote more?

I'd say thats architecture, i doubt you would though or is the point where you backtrack and spout some other "factual" information?

Deadlocks can be avoided in any database system by coding properly.

-- 
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials


"Galen Boyer" <galen_boyer_at_yahoo.com> wrote in message 
news:upslkh02y.fsf_at_rcn.com...


> On Sat, 18 Feb 2006, tonyrogerson_at_sqlserverfaq.com wrote:

>>> How does this lead credence to the fact the SQLServer's concurrency

>>> model is somehow on par with Oracle's?

>>

>> I didn't realise I was comparing, I thought I was pointing out Mark

>> Townsend floors in what he stated that Oracle takes care of everything

>> - well it doesn't! You still as a developer need to code for

>> concurrency, deadlocks etc... so trying to flower it up into some

>> magical beast is laughable.

>>

>> Do a google on oracle and deadlock and there is tons of stuff so don't

>> try and pull wool over my eyes. You do get deadlocks in Oracle and the

>> reasons give in the articles I've read are exactly the same reasons we

>> get deadlocks in SQL Server.

>

> Please explain.  A deadlock in Oracle is an application bug, only.  A

> deadlock in SQLServer is something that cannot be avoided.  They can be

> avoided in Oracle.  I'm not pulling the wool over anyone's eyes.

> Deadlocks happen in Oracle, but because of poor programming, not poor

> underlying server architecture, contrary to SQLServer.

>

>> I don't understand why you are so touchy on this subject....

>

> Touchy?  Not in the least.

>

> -- 

> Galen Boyer