Dec. 2 — The 4th HPC Knowledge Meeting, organised by HPCNow! and IQTC-UB, will take place in Barcelona, Spain, from February 2nd to 6th, 2015. HPC Knowledge Meeting is a key global conference for High Performance Computing, Clustering, and High Performance Data Analytics. Up to 250 HPC specialists are expected throughout the week, an even more extensive conference program than previous years.
BeeGFS Training on Monday, February 2
BeeGFS (formerly known as FhGFS) is the high-performance parallel file system from the Fraunhofer Competence Center for High Performance Computing. BeeGFS is provided free of charge. Professional commercial support is available directly from Fraunhofer (through ThinkParQ GmbH) or from international partners like HPCNow!
The BeeGFS Training’15, organised by HPCNow!, IQTC-UB and ThinkParQ, will be held at the University of Barcelona, on February 2, 2015, in conjunction with the HPC Knowledge Meeting’15. This training will provide key knowledge in order to take advantage of the features and components of this Clustered Filesystem like distributed and mirrored metadata architecture, high availability or multi-tier storage level.
The agenda is already available on the HPC Knowledge Portal (www.hpckp.org).
Slurm Workload Manager Training on Thursday and Friday, February 5 and 6
Slurm Administrator Training’15, organized by HPCNow! and IQTC-UB, is aimed to share expertise and strategies of the most popular Batch Queue System in High Performance Computing. It will be held at the University of Barcelona, on February 5 and 6, 2015, in conjunction with HPC Knowledge Meeting’15.
The training materials will be developed and peer-reviewed by active members of the HPC community with deep experience in Slurm Workload Manager. The training materials will be available as well as the previous Grid Engine trainings.
The Call for Presentation & Tutorial Proposals is already open. Experienced Slurm users are encouraged to submit proposals for Presentations and Tutorials to be given during the training course. Concurrent sessions are scheduled in 50-minute and 2-hour blocks. We would like to cover all areas of interest of Slurm Workload Manager. For that reason, we encourage you to submit presentations or tutorials related with this software. Otherwise, you can also contribute by providing feedback, suggesting new topics, improving the quality of the training material by peer-review process or helping other attendees of this training.
HPC Knowledge Meeting Highlights, February 2 to 6, 2015
This year, a major focus will be given to convergence of the HPC to the Big Data. In recent years, the development of a new paradigm of computing, as was the appearance on the scene of Big Data model, has allowed a lot of economic sectors to take advantage of the use of supercomputing. Big Data can work with huge volumes of information quickly and efficiently, and at the confluence with the world of supercomputing applications HPDA (High Performance Data Analysis) was born. These applications perform tasks with a volume of data and algorithmic complexity such that require very large computational resources.
Finally, since we remain faithful to our origins, several sessions will be devoted to classical HPC and clustering.
Call for Presentation & Tutorial Proposals is now open. HPC & clustering experts are encouraged to submit proposals for Presentations and Tutorials to be given during the conference. Concurrent sessions are scheduled in 15- minute, 30-minute and 60-minute blocks. We would like to cover all areas of interest of HPC and clustering, with special interest in High Performance Data Analytics.
Attendees of the HPC Knowledge Meeting and satellite events should expect five intensive days of pure community knowledge sharing and technology transferring.
HPCKP wants to acknowledge again the sponsors, Fujitsu, HPCNow!, ThinkParQ, XRQTC, Omega Peripherals, and SIE for the valuable contribution in this event.
Further information: www.hpckp.org
—
Source: HPCKP