There are three simple methods of estimating capacity requirements for Pervasive PSQL Vx Server:
- Download a trial of PSQL Vx Server, run the application and review the data in the Monitor utility.
- If you are running PSQL v11, update your application to PSQL v11 SP2 and review the data in the Monitor utility.
- Use Monitor data from a production system with PSQL Server or Workgroup to provide a capacity estimate.
In all cases, it is very important to collect data from a system that is running at or near the application's peak load. It is also important to identify seasonal or cyclical peaks and make a decision about purchasing permanent or time-limited capacity to meet the occasional increase in demand.
Estimating Capacity Requirements Using Pervasive PSQL Vx Server or PSQL v11 Server SP2
The best way to get a clear picture of an application's PSQL Vx Server capacity requirements is to:
- Download a trial copy of PSQL Vx Server or update your PSQL v11 application to SP2
- Run the application under conditions that simulate a full load
- Review data collected by the Monitor utility
- Look at the table below for the appropriate PSQL Vx Server size
Pervasive PSQL Vx Server has an updated version of the Monitor utility that will track peak values for Data in Use and Session Count for as long as the engine has been running. To view the peak values, launch the Monitor utility an display Microkernel Resource Usage. From the screenshot below, one can see that the peak value for Session Count is 2 and the peak value for Data in Use is 2 MB.
In order to get an accurate estimate of PSQL Vx Server capacity requirements, it is very important during the test to simulate a production at full load for a defined number of users. This will help when it comes time to translate the load for a certain number of users to a capacity requirement for Pervasive PSQL Vx Server.
Estimating Capacity Requirements with PSQL Monitor from PSQL v11 SP1 or Earlier
Production systems can be used with versions of Monitor from PSQL v11 Service Pack 1 or earlier to provide some insight into the application's PSQL Vx Server capacity requirements. Here's a quick summary of how this works:
- Run the application with full load for a typical number of users and for enough time to ensure peak resource usage.
- Open Monitor utility and review Peak Clients in MicroKernel Resource Usage
- Using MicroKernel Active Files view, identify directories with opened database files
- Add up the total file sizes for those directories
- Look at the table below for the appropriate PSQL Vx Server size
- Remember: If your application requirements experience cyclical or seasonal peak loads, simulate these and include the numbers in your estimates.
Microkernel Resource Usage in these versions of Monitor does include a peak number for Clients (the closest analog to Sessions). However, because the Clients metric includes connections used for internal database processes (for example, accessing PSQL system files) and these are not included in Session Count, the number for Clients will always be higher than Session Count.
The MicroKernel Active Files table includes a list of directories containing files that have been opened by the database. Calculating the total size of the files in these directories will provide an outside estimate of Data in Use requirements for Pervasive PSQL Vx Server. This is because Data in Use only includes opened files and does not include files opened for internal database processes. Adding up the total size of files in directories listed in the MicroKernel Active Files table will typically deliver an inflated approximation of Data in
Use.
Pervasive PSQL Vx Server - Standard Capacities
| Small |
25 |
5 GB |
| Medium |
100 |
20 GB |
| Large |
250 |
50 GB |
| SuperSize |
Unlimited |
Unlimited |
Additional Resources