Astro-HPC 2012 - Workshop on High-Performance Computing for Astronomy

In the past 5 years, astronomy has become one of the biggest consumers of computing resources. Therefore, new computational solutions (both hardware and software) are emerging, dedicated to the various fields of the science. For example, in radioastronomy, instruments are becoming extremely large. Comprising of thousands of antennas, these radio telescopes generate huge amounts of data that have to be analyzed in a timely manner. As a result, large scale systems (both distributed and centralized) are employed for data gathering, filtering, analysis, and imaging. Furthermore, the Square Kilometer Array (SKA), a very large scale international project, seems to push the limits of radioastronomy data processing to new limits, as proven by pathfinders such as LOFAR, ASKAP, MeerKAT or anzSKA.
All in all, it is expected that astronomy experiments will become larger in all dimensions: larger data collections, more accurate data analysis and processing, and more detailed results (i.e., imaging). This significant increase of the large-size experiments to be performed by the instruments built around the world require not only huge processing power, but also clever system design, all the way from the hardware construction to the software development and deployment.
In this context, this first edition of the Astro-HPC workshop focuses on system design for large-scale astronomy systems. We aim to give specialists from both astronomy and computer science and engineering the opportunity to discuss both the requirements of large-scale high-performance computing systems suitable and/or usable for various astronomy applications, and practical examples of such designs and implementations. Therefore, we encourage both contributions that analyze the size and needs of large-scale astronomy systems to be used in the near future, as well as contributions that show how existing algorithms and methods should be adapted or replaced to the larger scale of these near future experiments.
The First Workshop on Social Media Processing (SocMP’12) solicits papers that address the data management and processing challenges raised by running and mining large-scale social media platforms from a systems and infrastructure-oriented perspective. The workshop will be co-located with ACM/IEEE HPDC (http://www.hpdc.org/2012/) and will take place in June 2012 in Delft, The Netherlands. SocMP’12 will bring together researchers and practitioners in discussing and creating new knowledge about the social media infrastructure and methods of the future.
Topics of interest include, but are not limited to:
Systems
Specific design requirements and solutions
Benchmarking and performance evaluation
Performance analysis and limitations of new system configurations
Energy efficiency: predictions, measurements, analysis.
Accelerators
Evaluation of accelerator-based systems in the field of astronomy
The requirements for accelerators in astronomy applications
The impact of various accelerator on the performance and/or energy of HPC systems for astronomy
Applications
Novel large-scale astronomy applications
Evaluation and benchmarking of applications in large-scale environments
Updates, tuning, and optimizations of existing astronomy applications
Algorithms
Performance evaluation and analysis of traditional algorithms
Novel algorithms for astronomy kernels