Companion of the SimGrid storage modeling article

The tremendous increase in scientific data production and the<br>ever-growing need for data analysis and preservation coming from<br>various scientific domains create a great emphasis on storage<br>components. Understanding the performance of a storage subsystem or<br>dimensioning it properly is an important concern independent of the<br>scale and type of the associate computing infrastructure. Data<br>centers, Supercomputers, Grids, and Clouds comprise storage components<br>whose specifics may differ but all need to be well understood. The<br>study of such systems often mandates empirical evaluation of solutions<br>via simulation. Unlike direct experimentation, simulation enables<br>fully repeatable and configurable experiments that can often be<br>conducted quickly for arbitrary hypothetical scenarios. However, most<br>simulation frameworks tailored for the study of distributed<br>systems offer no or little abstractions of storage infrastructures.<br><br>In this paper, we present our efforts to extend the simulation<br>capacities of the \simgrid framework with abstractions, models, and<br>interfaces that allow users to manipulate storage components and their<br>contents in their simulations. We detail our design choices, the<br>proposed implementation, and experiments designed to <br>study the validity of the proposed model.<br><br><br>