Parallelizing dense matrix computations to distributed memory architectures is a well-studied subject and generally considered to be among the best understood domains of parallel computing. Two… (More)
To implement dense linear algebra algorithms for distributed-memory computers, an expert applies knowledge of the domain, the target architecture, and how to parallelize common operations. This is… (More)
We show how transformations organize and explain the designs of legacy pipe-and-filter-architectures. We start with an elementary architecture and progressively transform it to a detailed executable… (More)
BACKGROUND
Human emotion is a crucial component of drug abuse and addiction. Ultrasonic vocalizations (USVs) elicited by rodents are a highly translational animal model of emotion in drug abuse… (More)
BLIS is a new software framework for instantiating high-performance BLAS-like dense linear algebra libraries. We demonstrate how BLIS acts as a productivity multiplier by using it to implement the… (More)
Design by Transformation (DxT) is a top-down approach to mechanically derive high-performance algorithms for dense linear algebra. We use DxT to derive the implementation of a representative matrix… (More)
DxTer is a tool that generates a search space of implementations for operations in dense linear algebra and uses cost functions to select automatically the most efficient implementation from the… (More)
Dense linear algebra (DLA) algorithms for distributed memory architectures are often implemented as sequences of highly optimized parallel implementations of individual sub-operations. This can… (More)
We show empirically that some of the issues that affected the design of linear algebra libraries for distributed memory architectures will also likely affect such libraries for shared memory… (More)
Design by Transformation (DxT) is an approach to software development that encodes domain-specific programs as graphs and expert design knowledge as graph transformations. The goal of DxT is to… (More)