Fortran is dead – Long live Fortran!

Torsten Hoefler's random access spontaneous talk given at the 42nd anniversary Salishan Conference on High-Speed Computing in 2023. Discusses how to lift Fortran code to a data-centric representation to optimize it for accelerator devices. Work led by Alexandru Calotoiu in SPCL.

[Machineman]

Fortran is still used today, particularly in scientific, engineering, and high-performance computing applications where numerical computation and performance are critical. While newer languages like Python and Julia have gained popularity for general-purpose programming and rapid prototyping, Fortran remains widely used in fields such as computational physics, climate modeling, computational chemistry, and finite element analysis

[ChrisPollitt]

TIOBE Index for May 2024: Fortran in the top 10

[kamertonaudiophileplayer]

My friend claimed that he can program in Fortran everything. How it is true! I also converted many original weather model calculations from Algol to Fortran. They work great.

[patrickkearney]

First two not verbatim quotes, real programmers can write FORTRAN code in any language and computer scientists solve yesterday's problems with tomorrow's hardware.
i am as old as FORTRAN and have coded in FORTRAN, C, APL ALGOL, Forth, LISP, Basic, Pascal, Mathematica, MATLAB, various scripting languages, ... even run scientific computation on laser printers overnight using POSTSCRIPT. Also I have written real time operating systems in C and assembly language.
I firmly believe that the design philosophy of modern computer languages became decoupled from consideration of current and future hardware capabilities and the more general resources available prior and during execution of a program. Also operating system support for client processes is mostly very poor.
For example, parallel code execution was not possible on early computers. Modern vector processors, FPGAs or multi core CPUs can handle concurrent parallel computation but efficient formalized code design is sorely lacking.

[jameschums]

thank you for introducing some concepts i had not really considered before. will future compiler mix languages and optimise codes to use different hardware for operations. great talk, thank you, now i am thinking AI tools could optimise fortran, c/cuda, python... ?