Authors: Philipp Falk (ThinkParQ, Germany), Manisha Salve (ThinkParQ, Germany), Rui Màrio Silva Machado (Fraunhofer Institute for Industrial Mathematics)
Abstract: BeeGFS is an award-winning parallel file system designed for performance-oriented environments, widely seen by the HPC community as a robust, easily-deployable and scalable solution compared to alternate parallel file systems. BeeGFS is created on an “Available Source” development model (source code is publicly available), offering a self-supported Community edition and a fully supported Enterprise edition that includes additional features and functionalities.
This session brings together the BeeGFS and HPC communities, with the architecture team behind BeeGFS to connect, interact and discuss the product roadmap and development plans, and share its users' experiences, including how BeeGFS accelerates their HPC infrastructures.
Long Description: The session will be led by members of the BeeGFS community along with BeeGFS developers who will openly discuss the challenges of parallel file systems along with how a parallel file system accelerates their HPC infrastructure along with the benchmarking and testing they undertook in the selection process of finding the right solution. Along with outlining the product roadmap and development plan with the community, the BoF session will also bring the community together to discuss the future development plans of BeeGFS. This open discussion will help prioritize what features BeeGFS users and the community requires in future product releases. The audience will be encouraged by the BeeGFS development team to interact throughout this session by sharing their views and feedback in relation to the product roadmap, direction, and development plans. This will provide the architectural and development team behind BeeGFS with insight into what features the community wants and how BeeGFS can be improved going forward. A BeeGFS user will provide a live use case of their recent deployment of BeeGFS and how a parallel file system accelerates their HPC system along with benchmarks and the process they followed during their selection process.
Back to Birds of a Feather Archive Listing