For more details see the (experiment plan)[TestPlan.url]
To perform the benchmarks you need to have a linux system with:
- make
- curl
- git
To compile the workload generator, you need:
- git
- go-lang 1.12
- make
To perform data analytics you need:
- python
- jupyter-notebooks
- scipy
It is recommended that you run scripts/prepare.sh once to set linux properties to perform this benchmark in a fair way.
To run the test modify run make bench-<location>, before that make sure to prepare the system under test according to the experiment plan.
Once your down with the experiments, run git add data/ and git commit -m "<Your Name> <Experiment Location>" && git push to upload the data to this repo.
Run make start to work on processing the data. Put all plots into the images folder.