Title :
Offline library adaptation using automatically generated heuristics
Author :
De Mesmay, Frédéric ; Voronenko, Yevgen ; Püschel, Markus
Author_Institution :
Dept. of Electr. & Comput. Eng., Carnegie Mellon Univ., Pittsburgh, PA, USA
Abstract :
Automatic tuning has emerged as a solution to provide high-performance libraries for fast changing, increasingly complex computer architectures. We distinguish offline adaptation (e.g., in ATLAS) that is performed during installation without the full problem description from online adaptation (e.g., in FFTW) that is performed at runtime. Offline adaptive libraries are simpler to use, but, unfortunately, writing the adaptation heuristics that power them is a daunting task. The overhead of online adaptive libraries, on the other hand, makes them unsuitable for a number of applications. In this paper, we propose to automatically generate heuristics in the form of decision trees using a statistical classifier, effectively converting an online adaptive library into an offline one. As testbed we use Spiral-generated adaptive transform libraries for current multicores with vector extensions. We show that replacing the online search with generated decision trees maintains a performance competitive with vendor libraries while allowing for a simpler interface and reduced computation overhead.
Keywords :
decision trees; learning (artificial intelligence); pattern classification; software libraries; statistical analysis; ATLAS; FFTW; Spiral-generated adaptive transform libraries; automatic tuning; automatically generated heuristics; decision trees; high-performance libraries; offline adaptive libraries; statistical classifier; Classification tree analysis; Computer architecture; Decision trees; Discrete Fourier transforms; High performance computing; Linear algebra; Runtime; Software libraries; Timing; Writing; FFT; automatic performance tuning; decision trees; fast Fourier transform; high-performance computing; library generation; machine learning; statistical classifier;
Conference_Titel :
Parallel & Distributed Processing (IPDPS), 2010 IEEE International Symposium on
Conference_Location :
Atlanta, GA
Print_ISBN :
978-1-4244-6442-5
DOI :
10.1109/IPDPS.2010.5470479