Test data vol inconsistent - on diff tables. How to extrapolate that?
I have an issue with my test data, i.e. we use 20 tables all together, but we only have 15 tables with 100% of test data and the rest of them (5 tables) have only 5% of test data volume. Due to time constraints we have to proceed with the test execution.
Since, the data volume on different tables is inconsistent. How do we extrapolate for 100% of data? and what r the things that i need to keep in mind during test execution due to this risk?
Re: Test data vol inconsistent - on diff tables. How to extrapolate that?
Have you analyzed how these 5 small size tables are used? Because if their usage is just occasional, the risk of your approach will be significantly lower than when these tables are used all the time.
When you have found out how and where these tables are used you can review if the indices are consistently used, if the tables participate in joins etc. based on this result you can further refine your risk assesment.
On your question if you can extrapolate: No you can't, but during your tests you can develop an understanding for the usage of the database and potential impacts when you have full tables in.