SPIRAL’s Operator Language: From Textbook Math to High Performance –With Correctness Guarantees
, Carnegie Mellon University
Date: Monday, September 21, 2020
Time: 2:00 PM to 3:00 PM Note: all times are in the Eastern Time Zone
Location: https://mit.zoom.us/meeting/register/tJUrdOqopj8uHdO4gUyVMnfglOFEqIye_Je0 (Registration required)
Event Type: Seminar
Room Description: https://mit.zoom.us/meeting/register/tJUrdOqopj8uHdO4gUyVMnfglOFEqIye_Je0 (Registration required)
Host: Julian Shun, MIT CSAIL
Contact: Julian Shun, firstname.lastname@example.org, email@example.com
Relevant URL: http://fast-code.csail.mit.edu/
Speaker URL: https://users.ece.cmu.edu/~franzf/
firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com
TALK: SPIRAL’s Operator Language: From Textbook Math to High Performance –With Correctness Guarantees
This talk discusses the Operator Language, the enabling technology in the SPIRAL code generation and automatic performance tuning system. We develop a formal framework to capture computational algorithms, computing platforms, and program transformations of interest, using a unifying mathematical formalism we call operator language (OL). Then we cast the problem of synthesizing highly optimized computational kernels for a given machine as a strongly constrained optimization problem that is solved by a multi-stage rewriting system. Since all rewrite steps are semantics preserving identity operations, our approach allows us to formally prove the equivalence between the kernel specification and the synthesized program.
Finally we present a first look at FFTX and SpectralPack. We aim at translating the LAPACK/BLAS approach from the numerical linear algebra world to the spectral algorithm domain. FFTX is extending and updating FFTW for the exascale era and beyond while providing backwards compatibility. SpectralPack captures higher level spectral algorithms and their variants, including convolutions, Poisson solvers, correlations, and numerical differentiation approaches that translate to FFT calls. The SPIRAL system--now available as open source under a BSD/Apache license--underpins the effort to provide performance portability.
Franz Franchetti is a full Professor in the Department of Electrical and Computer Engineering at Carnegie Mellon University. He received the Dipl.-Ing. (M.Sc.) degree in Technical Mathematics and the Dr. techn. (Ph.D.) degree in Computational Mathematics from the Vienna University of Technology in 2000 and 2003, respectively. Dr. Franchetti's research focuses on automatic performance tuning and program generation for emerging parallel platforms and algorithm/hardware co-synthesis. Within the Spiral effort, his research goal is to enable automatic generation of highly optimized software libraries for important kernel functionality.
Computer Architecture, Programming Languages & Software
Created by Julian J. Shun at Monday, September 14, 2020 at 11:23 AM.