Universidad Politcnica de Baja California Juan P Navarro
Universidad Politécnica de Baja California. Juan P. Navarro Sanchez 9 th level English Teacher: Alejandra Acosta The Beowulf Project.
My Proposal > This project consists of having the students from the university build a Beowulf style cluster > Can be done with available hardware > There is a wide range of open source software for this. > Implementation the basics of networking, and systems administration 2
The Background > Beowulf cluster > Built in 1994 by Thomas Sterling and Donald Becker for the NASA. > Named after an English poem. > A Beowulf cluster is a group of what are normally identical, commercially available computers, which are running a Free and Open Source Software 3
Cluster > A computer cluster is a group of linked computers, working together closely thus in many respects forming a single computer. 4
HPC > High-performance computing uses supercomputers and computer clusters to solve advanced computation problems. 5
The Grid > Grid computing is a term referring to the combination of computer resources from multiple administrative domains to reach a common goal. 6
TSUBAME: The inspiration for this project 7
8
9
Why the cluster? Energy Aerospace research Investigation > Unique 3 D Scientific modeling in research the area Cost effective 10
> > > Scientific research Energy research Virtualization 3 D modeling Aerospace investigation Others 11
The software > Message Passing Interface (MPI) programming model; > Parallel Virtual Machine (PVM) programming, execution, and debugging model; > Parallel file system; > Tools to configure, schedule, manage, and tune parallel applications; > Higher level libraries. 12
13
The hardware > Switches > Cable > Cluster servers (these are the ones doing all the processing) > Cluster head (scheduling tasks, managing the whole cluster) > Host, user 14
15
NVidia Tesla Cards > High parallel capabilities > Low power consumption > Inexpensive compared to other solutions 16
Programming > Students from UPBC can work developing part of the software needed like Matsuoka Satoshi did with the TSUBAME grid. 17
Conclusion > This cluster if ever done would be of great help as it would let us do research and make advances to the technology. > Having this cluster is necessary if the university plans to grow bigger. > Finally the project should be done by the students with supervision from the school teachers, or work side by side with the engineers of the project. 18
REFERENCES > Lou, C. , Wu, L. , Yi, O. (2002). Application of parallel virtual machine framework to the strong prime problem. Journal of Computer Math, 79(7), 797– 806. doi: 10. 1080/ 00207160290029228 > Banerjee, S. , Mukherjee, I. , Manhati, P. (2009). Cloud Computing Initiative using Modified Ant Colony Framework. World Academy of Science. 56(2), 221 -224. > Leigh, J. , Brown, M. (2008). Merging Real and Virtual Worlds. Comunications of the ACM, 51(1), 82 -85. > Axelsson S. (2002). Intrusion detection systems: A survey and taxonomy. Technical Report, 18, 231 -241 19
> Enlanger, L. , Lackow, A. (2006). High-Performance Computing: Supercharging the Enterprise. Retrieved from http: //www. infoworld. com/ 20 -23. > Linz, J. (2001). Analysis on Supercomputers and Clusters. World Scientific Publishing. 11(2 -3), 252 -262. > Clement, T. (2009). Unleashing the Potential of High. Performance Clusters. Retrieved from http: //www. scientificcomputing. com/ 18 -22. > Bell, G. , Gray, J. (2002). What’s Next in High-Performance Computing? . Communications of the ACM. 45(2), 91 -95. 20
> Tokyo Institute of Technology Global Scientific Information and Computing Center. (2010). What is STUBAME. Retrieved from http: //www. gsic. titech. ac. jp/en/tsubame > The STUBAME Grid (2006). Redefining Supercomputing. Retrieved from http: //www. sun. com/servers/x 64/x 4600/tsubame. pdf 1 -4. > High Performance Computing (2011). Supercomputing with Tesla GPUs. Retrieved from http: //www. nvidia. com/object/tesla_computing_solutions. h tml 21
Thank you 22
- Slides: 22