Developer Documentation Library > Interviewer - Server > Monitoring and tuning system performance > Recommendations and points to consider
 
Recommendations and points to consider
Recommendations
If an interviewing server has more than 2Gb of memory, you can configure multiple interviewer session engines on that computer. The recommendation is one engine per 1Gb of RAM. See “Adding session engines to an interviewing server” in the UNICOM Intelligence Interviewer installation instructions for details.
The recommendation for cluster installations is three web servers for every four interviewing servers.
Points to consider
How many projects will run concurrently? Memory is less of an issue when running fewer projects.
How big are those projects?
How complex are the questions: for example, all simple categorical or numeric questions, or lots of large loops, grids, blocks, or compounds? lots of complex routing?
How many templates do you use, and how complex are they?
Do you use images? If so, how many and how big are they?
Do your projects use sample management?
Do your projects use quota control?
See also
Monitoring and tuning system performance