My opinion on MTS is that unless you really need it, you should avoid it. Running MTS adds at least two extra IPC latencies to every call and some CPU usage. It also introduces some extra points of potential contention, some of which were mentioned in yesterday's post on Tuning shared servers.

You should only use MTS if your application is using technology that mandates it (like EJBs) or if you have definitely reached your scalability limits it terms of process counts or possibly virtual memory. With MTS you need fewer processes, thus context switching is cheaper and the total virtual memory size is a little lower (but the SGA is much larger). So MTS can let you support more users on a system that is already running at capacity, but increasing that capacity is normally a better solution.

I have a question on MTS server model. At our client site, concurrent access to database are going to increase from 200 (as of now) to 1200 users (in the future). We are thinking of implementing MTS database setup to handle the concurrent access, work load and to improve the performance.

Before deciding this solution, I would like to know is there any known/unknown problems with MTS database setup.