Deinsum: Practically I/O Optimal Multi-Linear Algebra for Distributed Systems
Offered By: Scalable Parallel Computing Lab, SPCL @ ETH Zurich via YouTube
Course Description
Overview
Explore a conference talk on Deinsum, a framework for distributed multilinear algebra computations using Einstein notation. Delve into the challenges of optimizing data movement in massively-parallel systems and learn how Deinsum addresses these issues. Discover the framework's approach to deriving data movement-optimal tiling and generating distributed schedules, enhancing local computation performance. Examine the application of Deinsum to important tensor kernel classes, including Matricized Tensor Times Khatri-Rao Products and Tensor Times Matrix chains. Gain insights into the framework's workflow, binary operations, communication minimization techniques, and optimal tile size determination. Analyze experimental results showcasing Deinsum's performance on the Piz Daint supercomputer, demonstrating significant speedups over existing solutions.
Syllabus
Introduction
Modern Scientific Applications
Deinsum Workflow
Binary Operations
Minimize Communication
Example
Cartesian Process Grid
Optimal Tile Sizes
Annotation
Data Distribution
Code Generation
Experimental Results
Conclusion
Taught by
Scalable Parallel Computing Lab, SPCL @ ETH Zurich
Related Courses
Intro to Parallel ProgrammingNvidia via Udacity Introduction to Linear Models and Matrix Algebra
Harvard University via edX Введение в параллельное программирование с использованием OpenMP и MPI
Tomsk State University via Coursera Supercomputing
Partnership for Advanced Computing in Europe via FutureLearn Fundamentals of Parallelism on Intel Architecture
Intel via Coursera