Use case class 3 -- User Story

Use case class 3:

Present James:

James wants to simulate how a molecular cloud collapses and forms a cluster of young stars. These simulations can, in principle, be performed with publically available codes. However, they are computationally costly, so that only a single simulation is performed. However, this means that the result might not be statistically representative. Thus James would like to apply for computer time at all the different supercomputers in Germany. She also would want to publish the results so that the investment into computer time pays off as others can use the results for their research either by analyzing them differently, using the results as input data, or comparing them to observations. However, it turns out that James would have to adapt the code to every single HPC computer, which would take several weeks of work for each of the new computational setups. Besides, she does not know where to store the data so that they are publically available and which information needs to be provided that others can reuse the data.

Future James:

James can use the code straightway as the code is already optimized for use at the HPC centers. Not only that but even if the HPC center offers its services on a new computer, the code will be adapted to this new machine. Now she can obtain statistically reliable results. She also does not have to care about where to store the data she wants to publish.  This adjustment is automatically done for her. She only has to choose which of her results are the ones she uses for her publication. Then the data will be stored easily identifiable by a DOI. Besides, the input setup will also be stored and made publicly available as the initial conditions and simulation parameters will be published with the data. She can choose whether she wants to directly interact with the different HPC centers, or for convenience do it through PUNCH-SDP.