Finalize testing plan

[YaphetKG]

Next step finalizing testing plan is to get some of the sample queries as test cases, Also including plan on configuration of test including things such as

• Rate of traffic
• Frequency of test runs Living document (https://docs.google.com/document/d/1mvmPPrHM8wklAlokJbU2f5ncW0EWAuBBIWjRRxpQ9vI/edit#heading=h.gxn55sko4myx) Once document is finalized plan is to add these into predator instance .

[YaphetKG]

Testing plan is ready (https://docs.google.com/document/d/1mvmPPrHM8wklAlokJbU2f5ncW0EWAuBBIWjRRxpQ9vI/edit#heading=h.o45zfz7vq5zp)

[cbizon]

Are we ready to implement the plan now?

[YaphetKG]

We now have a first round of tests on the predator server (https://predator-load-testing.apps.renci.org/ui/tests) The tests are the cartesian of this queries (https://github.com/frink-okn/api-tests/blob/main/performance-tests/tests.json) with this set of servers (https://github.com/frink-okn/api-tests/blob/main/sparql-servers.json) . This (https://github.com/frink-okn/api-tests/blob/main/create_artillery_tests.py) script can be used to update the tests.

Next steps:

1. The performance tests might need a more closer look , this are very simple queries , and some of them are most likely going to require parametrization. (@mahir256 , @balhoff ) if you look at them and provide any feedback on these that would be awesome. (https://github.com/frink-okn/api-tests/blob/main/performance-tests/tests.json)
2. Think about how to schedule these.
3. Think about setting benchmarks, (running these and settling on a performance we like) and setup notification system, if scheduled runs fall behind the benchmarks (?)
4. Add tests with regards to Regression as fore mentioned in the testing plan.