General validation of LAPIS

[eileen-kuehn]

For publications and so on we need a validation of the simulation approach of LAPIS.

I do see some issues here as we deal with opportunistic resources, meaning, that the environment is dynamic. It is, therefore, hard to monitor a real system as there are too many variables that have influence on the actual behaviour of the system. I, therefore, propose to first validate the static setup of simulation.

In a first step I would like to set up a demonstration to compare our simulation results to results from other simulators e.g. Alea.

In preparation for this ticket we should take care on:

• with which measurements / based on which concepts do other simulators validate their results,
• what kind of plots should we use to validate the simulation,
• what input data do we need? Do we already have the required information?
• ...

Plots

• Performance Benchmarking, also relevant for Benchmark for simulation with/without caching #70