Auto-Tuning the Matrix Powers Kernel with SEJITS

TitleAuto-Tuning the Matrix Powers Kernel with SEJITS
Publication TypeConference Paper
Year of Publication2012
AuthorsMorlan, J., Kamil S., & Fox A.

The matrix powers kernel, used in communication-avoiding
Krylov subspace methods, requires runtime auto-tuning for best performance.
We demonstrate how the SEJITS (Selective Embedded Just-In-
Time Specialization) approach can be used to deliver a high-performance
and performance-portable implementation of the matrix powers kernel
to application authors, while separating their high-level concerns from
those of auto-tuner implementers involving low-level optimizations. The
bene ts of delivering this kernel in the form of a specializer, rather than a
traditional library, are discussed. Performance of the matrix powers kernel
specializer is evaluated in the context of a communication-avoiding
conjugate gradient (CA-CG) solver, which compares favorably to traditional

Auto tunning the matrix powers.pdf351.36 KB