Mark Schmalz (University of Florida)
Keywords: Astronomical Imaging, Image and Signal Processing, High-Performance Computing
Abstract:
Increasing image and signal data burden associated with astronomical image processing in support of space situational awareness implies much-needed growth of computational throughput beyond petascale (1015 FLOP/s) to exascale regimes (1018 FLOP/s, 1018 bytes of memory, 1018 disks and Input/Output (I/O) channels, etc.) In addition to growth in applications data burden and diversity, the breadth and diversity of high performance computing architectures and their various organizations have confounded the development of a single, unifying, practicable model of parallel computation. Therefore, models for parallel Exa Scale processing have leveraged architectural and structural idiosyncrasies, yielding potential misapplications. In response to this challenge, we have developed a concise, efficient computational paradigm and software called Parallel Computing with Exascale Mapping (PCEM) to facilitate efficient mapping of annotated application codes to parallel exascale processors.
Date of Conference: September 10-13, 2013
Track: Poster