CBM R&D: Scalable and Fault-Tolerant Cluster-Management for FLES

The cooperation project is working on numerous tasks for the design and implementation of the CBM experiment which is part of the FAIR particle accelerator. ZIB supports the design and implementation of a scalable, fault tolerant cluster architecture with proper cluster management and the efficient, distributed merging of the input data streams (approximately 1 Terabytes / s). Furthermore, we address the fault tolerant monitoring and scheduling of the online analysis cluster (FLES).

Publications

2019 2018 2017

2019
Farouk Salem, Florian Schintke, Thorsten Schütt, Alexander Reinefeld	Improving the throughput of a scalable FLESnet using the Data-Flow Scheduler	CBM Progress Report 2018, pp. 149-150, 2019, isbn: 978-3-9815227-6-1	BibTeX DOI
Farouk Salem, Thorsten Schütt, Florian Schintke, Alexander Reinefeld	Scheduling Data Streams for Low Latency and High Throughput on a Cray XC40 Using Libfabric	CUG Conference Proceedings, 2019	PDF BibTeX
Farouk Salem, Florian Schintke, Thorsten Schütt, Alexander Reinefeld	Scheduling data streams for low latency and high throughput on a Cray XC40 using Libfabric	Concurrency and Computation Practice and Experience, pp. 1-14, 2019	PDF BibTeX DOI
2018
Farouk Salem, Florian Schintke, Thorsten Schütt, Alexander Reinefeld	Data-flow scheduling for a scalable FLESnet	CBM Progress Report 2017, pp. 130-131, 2018, isbn: 978-3-9815227-5-4	BibTeX DOI
2017
T. Ablyazimov, A. Abuhoza, R. P. Adak, Alexander Reinefeld, Florian Schintke, et al.	Challenges in QCD matter physics – The scientific programme of the Compressed Baryonic Matter experiment at FAIR	The European Physical Journal A, 2017	BibTeX DOI
Farouk Salem, Florian Schintke, Thorsten Schütt, Alexander Reinefeld	Supporting various interconnects in FLESnet using Libfabric	CBM Progress Report 2016, pp. 159-160, 2017, isbn: 978-3-9815227-4-7	BibTeX