Auto-tuning parallel software: an interview with Thomas Fahringer the multicore transformation (ACM Ubiquity symposium)

Jul 7, 2014

In this interview conducted by Ubiquity editor Walter Tichy, Prof. Thomas Fahringer of the Institute of Computer Science, University of Innsbruck (Austria) discusses the difficulty in predicting the performance of parallel programs, and the subsequent popularity of auto-tuning to automate program optimization. The online link to the interview is here. The offline pdf-file is here. More

Distributed and Parallel Systems Group

DPS focuses on research that enables a high degree of easy-to-use and seamless automation for flexible collaborations and computations on a global scale. For our research we consider highly parallel and heterogeneous, dynamic, failure-prone distributed and parallel systems, services and resources. DPS enables new forms of science that are computation and data-intensive, distributed, collaborative, and multi-disciplinary. Our objective is to simplify the effective use of multiprocessor architectures and geographically distributed, heterogeneous systems, which will provide the basis for the next generation of Internet-enabled performance-oriented applications. DPS research should enable scientists to do faster, better or different research.

dps_triangle Green IT Green IT Parallel Processing Parallel Processing Cloud Computing Cloud Computing

Funded by: 

FWF Information Society Technologies Zukunft IFILandTirol



Part of  the: