Overall Considerations

To place the size of a system in perspective, it is worth looking at some statistics that show activities from existing installations of ExtraView. The example below shows the likely usage pattern for an organization of 100 users. For simplicity it is assumed that one report takes five times the CPU time and resource as an insert or update operation. The number of reports generated may seem high, but note that users will frequently visit their Home Page where there might be dashboards configured with several reports, as well as several individual reports existing on the Home Page.

Number of users 100
Average number of new problems entered per day per person 3
Average number of updates applied per day per person 10
Average number of reports run per day per person 50
A work day of 8 hours 8
Total number of complete database operations
   = 100 * ( 3 + 10 + ( 50 * 5 ))
26,300
Server operations per second
    = 26300 / ( 8 * 60 * 60 )
~ 0.9

Each of these operations may take about 1.5 seconds to complete on the CPU of a server. Although there will be peaks in the usage of the system, two processors could handle this size of load with reasonable results. More than two processors will give excellent results.

Note that these statistics are only an approximation and there are many other factors involved. It is not easy (or even possible) to judge the interaction of all these factors to come up with a precise performance metric.

Other factors that will affect the performance include:

  • Speed of the processors
  • Amount of memory within the servers
  • Amount of fields rendered within your ExtraView configuration's layouts
  • Amount of logic being processed with an issue, in the form of rules
  • The complexity of the reports being prepared
  • The effect of having one server or clustered servers
  • The available network bandwidth