Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> Re: Database or store to handle 30 Mb/sec and 40,000 inserts/sec
On Fri, 17 Feb 2006, tonyrogerson_at_sqlserverfaq.com wrote:
>> From what I've seen in my lab it is a pathetic implementation >> covering up for the inability of the engine's native capability with >> lots of CPU and lots of hard disk writes.
Here we have more tell-tale signs. Once again. You have to make sure one's transactions are short, or resource contention. Where, in any Oracle forum, do we recommend "shortening" one's transacations? The transaction handles the business need of the transaction. But, in SQLServer, just always be congnizant of the resource contention.
> you won't get disk activity because it will be done in cache.
>
> Does work though and scales.
As long as one is congnizant of the resources being used and coding accordingly.
> I'd be interested to see the SQL and schema you used on your tests -
> if you can send it me that would be great.
>
>> >> I think they went for the marketing claim ... and to satisfy SAP ... >> rather than actually fixing the underlying weakness in the design of >> the kernel.
You have got to be @#$%^ us? Having to resort to read uncommitted in the past? So, now, in this latest release, MS got it right? SQLServer has always made one accept lesser transactional integrity. Never, in Oracle, will you EVER, EVER be allowed, EVER, to see uncommitted transactions. This isolation level is for inferior database implentations, and even in the thread where you are touting the latest and greatest of SQLServer, you mention in passing one of the worst lines I've ever seen in a database group. It goes something like the following:
Q: I have query that is being blocked until a transaction on the table
it is accessing is complete.
A: (By a SQL Server MVP) Set transaction isolation level read
uncommitted.
Yep, this latest release fixes all of this. READ COMMITTED SNAPSHOT is its name. And, if it doesn't, all you need to do is code up some .net function and apply it to the CLR MLC XYZ engine.
-- Galen BoyerReceived on Fri Feb 17 2006 - 19:44:03 CST