Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore the evolution and future potential of Fortran programming language in this 44-minute conference talk that bridges historical context with cutting-edge applications in high-performance computing and artificial intelligence. Discover how Fortran 2023 can finally realize John Backus's functional programming vision from 1977, connecting the dots between the Harlem Renaissance, early computing history, and modern parallel programming challenges. Learn about innovative language facilities for asynchronously evaluating expressions that apply discrete, parallel, purely-functional differential operators to software abstractions modeling continuous mathematical concepts in partial differential equation solvers. Examine how Fortran's native concurrent loop iterations combine with side-effect-free, pure procedures to enable automatic parallelization of deep-learning inference and training algorithms across processors and accelerators. Gain insights into Berkeley Lab's Fortran team's multi-level software stack approach, spanning applications, compiler runtime libraries, and networking middleware, while understanding how inclusivity programs in science have facilitated significant aspects of this groundbreaking work in language-based parallel programming.
