Off-the-shelf real-time computers for next generation adaptive optics.pdf

Off-the-shelf real-time computers for next generation adaptive optics a b a Stefan Hippler , Douglas P. Looze , and Wolfgang Gaessler a ¨ ¨ Max-Planck-Institut fur Astronomie, Konigstuhl 17, 69117 Heidelberg, Germany bUniversity of Massachusetts, Amherst, Mass., U.S.A. ABSTRACT The performance of adaptive optics systems for existing as well as future giant telescopes heavily depends on the number of active wavefront compensating elements, the spatial, and the temporal sampling of the distorted incoming wavefront. In a phase-A study for an extreme adaptive optics system for the VLT (CHEOPS) as well as for LINC-NIRVANA a fizeau interferometer aboard LBT with a multi-conjugated adaptive optics system, we investigate how today’s off-the- shelf computers compare in terms of floating point computing power, memory bandwidth, input/output bandwidth and real-time behavior. We address questions like how level three cache can impact the memory bandwidth, what matrix- vector multiplication performance is achievable, and what can we learn from standard benchmarks running on different architectures. Keywords: Extreme Adaptive Optics, Adaptive Optics, Real-time Computer, COTS, MCAO, Control System 1. INTRODUCTION Existing adaptive optics (AO) systems used in astronomy are usually unique instruments designed for