JPlag Benchmarks

[sebinside]

The core comparison algorithm of JPlag is very efficient and can handle even larger tasks, e.g., dealing with 1000+ submissions.

We should demonstrate this better, i.e., by providing a public benchmark. This could include:

• [ ] A large dataset (one or more)
• [ ] A JPlag benchmark runner that takes the large dataset and measures the runtime and memory consumption
• [ ] A couple of numbers, potential dimensions are (choose two):
  • Different language modules (if relevant)
  • Different JPlag versions (if relevant)
  • Different systems (to showcase that JPlag runs on laptops and to demonstrate the parallelization)
  • ...?

The insights should at least be mentioned/linked or even be (partially) listed on the landing page / README.