Notícias

DEI TALKS | ''From Numerical Libraries, To Efficient Matrix Multiplication Compiler-Only Code Generation, To a Modular Automated General Packing Data Transformation''

17 de julho |11h30 | Sala B006 | FEUP

O Departamento de Engenharia Informática da Faculdade de Engenharia da U.Porto (FEUP) organiza, juntamente com o Núcleo Estudantil ACM FEUP, mais uma sessão das DEI Talks, em formato presencial, no dia 17 de julho pelas 11h30, na sala B006. “From Numerical Libraries, To Efficient Matrix Multiplication Compiler-Only Code Generation, To a Modular Automated General Packing Data Transformation” é a temática central da sessão proferida por J. Nelson Amaral, com moderação de Pedro Diniz, docente do DEI.

Nelson Amaral, a Computing Science professor at the University of Alberta with a Ph.D. from The University of Texas at Austin, has published in optimizing compilers and high-performance computing. Scientific community service includes general chair for the 23rd International Conference on Parallel Architectures and Compilation Techniques in 2014, for the International Conference on Performance Engineering in 2020, and for the International Conference on Parallel Processing in 2020. Accolades include ACM Distinguished Engineer, IBM Faculty Fellow, IBM Faculty Awards, IBM CAS “Team of the Year”, awards for excellence in teaching, the University of Alberta Graduate-Student Association Award for Excellence in Graduate Student Supervision, an University of Alberta Award for Outstanding Mentorship in Undergraduate Research & Creative Activities, and a recent University of Alberta 2020 COVID-19 Remote Teaching Award.

To support both Artificial Intelligence and High-Performance Computing workloads, new processors have introduced hardware acceleration for matrix multiplication. Examples include the Matrix Multiply Assist (MMA) in the IBM POWER10 and the Advanced Matrix Extensions (AMX) in the Intel Sapphire Rapids microarchitecture for Xeon servers. This talk describes how, in a collaboration between the University of Alberta, the University of Campinas, and IBM, we developed compiler technology to support such accelerators. An initial solution delivered a robust pattern matcher for General Matrix Multiplication (GEMM) computation operating at the LLVM intermediate representation that allows the replacement of the computation with an invocation of a high-performance library. A later solution delivered a compiler-only path for code generation by adapting the layered approach used in numerical libraries to the compiler code-generation process. Finally, a modular and automated general strategy for data packing, which can be applied to multiple algorithms, was developed for the Multi-Level Intermediate Representation (MLRI).

:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
Divulgue os seus eventos e projetos através do e-mail: noticias@fe.up.pt. Saiba mais sobre os nossos serviços no site dos SICC.