We are trying to create a performance testing strategy for Pega PRPC v6 to 7 upgrade.
Upgrading from Pega PRPC v6 to v7
For V7, we are planning to a fresh install and then migrate all data into v7 environment
The concern/issue is that Load testing simulation currently is done on a smaller subset of data (PT environment is hardly 10% in size and volume of Production). However, we are expecting to migrate over 10 Mn work objects after a planned upgrade. Their question is how do we make sure that the initial/baseline performance test done on a smaller subset scales for an environment with higher data set - basically how do we extrapolate ?
Also are there any best practices from performance testing perspective to be followed for this sceario?
Extrapolation is unreliable as you do not know the system pattern before hand (e.g., is it linear or nonlinear). Generally people try to simulate the Prod env by generating similar size of data and using tools like LoadRunner to mimic the actual load (e.g., number of users, etc) of the system. You probably can use the current available data size (10%) to run a series of load tests to gather some baseline data and go from there. The more tests you run, the more confidence you will have for the performance.