Call for Papers
2012 International Workshop on Modern Accelerator Technologies for GIScience
September 18, 2012. Columbus, Ohio, USA.
To be held in conjunction with the 7th International Conference on Geographic Information Science
(GIScience 2012 http://www.giscience.org)
GIScience community has been an active user of high-performance computing (HPC) resources both for data-intensive and compute-intensive applications. The HPC technology, however, is shifting from homogeneous systems employing identical processing elements to hybrid computing architectures that employ multi-core processors in combination with special-purpose chips. New multi-core architectures combined with application accelerators hold the promise of increasing performance by exploiting levels of parallelism not supported by the conventional systems. Examples of such architectures include Cell-based RoadRunner, hybrid system of combined CPUs and GPUs like Keeneland, and FPGA-based Novo-G systems. However, while there is general trend towards hybrid systems, current GIScience applications require significant algorithmic re-design and software re-engineering efforts to take advantage of the diverse computing resources.
The 2012 International Workshop on Modern Accelerator Technologies for GIScience will bring together developers of hybrid computing architectures and computing accelerators and experts involved with the development of applications on such systems to provide a collaborative forum for GIScientists, geographers, geoscientists, and computer scientists to discuss and share the state-of-the-art accelerator technologies and the pioneering applications, and promote collaboration on deploying the modern accelerator technologies in GIScience.
Topics of Interest
Topics of interest for MAT4GIScience 2012 workshop include two general categories. One is about how the accelerator technologies, such as GPUs, FPGAs, Multi-core & Parallel, and Multi-threading solutions, and heterogeneous computer architecture, are utilized in data and/or computation intensive GIScience applications, geospatial simulation and modeling, and remote sensing, as well as novel and parallel algorithms for geocomputation, spatial simulation and modeling, performance evaluation and optimization in high performance computing environment. The other is about the research challenges arising in GIScience, geography, geosciences, spatial simulation and modeling, and remote sensing that establishes the needs for high performance computing. By identifying the performance bottleneck in large scale geocomputation, the workshop promotes future collaboration on deploying modern accelerator technologies for high performance computing to advance fact-proven GIScience on the one hand, and advance the research in computational science and engineering, high performance computing, and geospatial algorithm development and optimization on the other.
Dr. David Bader, Georgia Institute of Technology.
Dr. Piyush Mehrotra, NASA Ames Research Center..
The workshop seeks the following three types of submissions related to MAT4GIScience 2012:
Research papers: up to 8 pages
Vision papers: up to 4 pages
Demo papers and short technical papers: up to 4 pages
Authors are invited to submit full, original, unpublished research papers that are not being considered for publication in any other place. Manuscripts should be submitted in PDF format and formatted using the guidelines provided by GIScience 2012. Papers must be submitted electronically, via the submission website on or before the submission deadline. All papers accepted before 07/21/2012 will be included in the GIScience 2012 proceedings.
Paper submission: 05/18/2012
Notification of acceptance: 06/22/2012
Camera-ready copies due: 07/06/2012
Workshop date: 09/18/2012
General Chairs and Contact Information
Xuan Shi, Georgia Institute of Technology. Email: email@example.com
Volodymyr Kindratenko, University of Illinois at Urbana-Champaign. Email: firstname.lastname@example.org