Can you really configure multiple database servers (all hosting exactly the same database) to carry out a single query concurrently?
I am not asking about performing queries using multiple Processor chips concurrently - I understand this it possible.
I mean , something similar to this:
- You will find two 2 servers: Server1 and Server2
- Both server host database Foo and both cases of Foo are the same
- I connect with Server1 and submit an elaborate (plenty of joins, many information) query
- Server1 decides that some information ought to be made on Server2 plus some data ought to be read from that server, too - appropriate areas of the query are delivered to Server2
- Both servers read data and perform necessary information
- Finally, is a result of Server1 and Server2 are incorporated and came back towards the client
All of this should happen instantly, without have to clearly reference Server1 or Server2. I am talking about such parallel query execution - is it feasible?
Just tips, John and wuputah.
I'm researching options of growing both availability and capacity of MOSS database after sales. What exactly I am searching for is a few kind out-of-the-box SQL Server load balancing solution that might be transparent towards the application, because I cannot customize the application by any means. I suppose SQL Server doesn't have such feature (and Oracle, so far as I realize it, does - it's RAC pointed out by wuputah).
An estimate in the Top Tips for SQL Server Clustering article:
Let us begin by debunking a typical misunderstanding. You utilize MSCS clustering for top availability, not for load balancing. Also, SQL Server doesn't have built-in, automatic load-balancing capacity. You need to load balance using your application's physical design.