FortranCon 2025

Nov 4, 2025, 9:00 AM → Nov 5, 2025, 10:00 PM Europe/Zurich

ZOOM (Virtual)

Alfio Lazzaro, Tiziano Müller

Synopsis

FortranCon 2025 is the third edition of the International Conference on the development and usage of the Fortran programming language, and it will take place on November 4-5, 2025.

With this conference, we intend to bring together active Fortran project developers from all fields: library and tool developers as well as people using Fortran to develop scientific applications, to share their experience and exchange ideas. The event is organized with contributions by the community for the community.

For the 2025 edition, the conference is being supported by members of the Back to the Fortran Future Network. The network is based in the UK, but with international reach, and includes representation from the UK, Germany, Netherlands, Sweden, France and the USA, and across academia, government, industry and other sectors (e.g., open source). Prof. Austen Rainer and Dr. Andrew Brown, both at Queen’s University Belfast in Northern Ireland, co-facilitate the network.

Go top of the page.

---

Keynote

For the Keynote talk we are happy to welcome John Reid, active member of ISO Fortran Committee ISO/IEC JTC1/SC22/WG5 (WG5 Convenor from 1999 through 2017).

Go top of the page.

---

Conference Structure / Registration / Call for Abstracts

The FortranCon 2025 is organized in two full days with session timings to accommodate attendance from both Europe and the Americas. Anyone interested is kindly requested to register for the event. The registration is free of charge. Please register by October 15th, 2025.  Participants will virtually attend the meeting. Instructions on how to connect to the event will only be sent to registered persons.

A Call-for-Abstracts is also open for people who would like to present at the conference. The official language is English. We have 2 tracks: Presentations (25 + 5 minutes discussion) and Communications (15 + 5 minutes discussion). They will be reviewed by a Scientific Committee.

The submission must consist of a title and an abstract (max 200 words). Furthermore, it is required to specify if you apply for a 15 or 25 minutes slot. You should submit a talk about anything you think is worth sharing with the Fortran Community. However, we are happy to host talks on other topics that can inspire us. Speakers will be requested to describe their experience, the implemented techniques, and the eventual new developments.

Unless explicitly requested by the speaker, all presentations will be recorded and made available through the FortranCon channel.

Submissions will be selected by review from the Scientific Committee via a voting process; members of the committee will not be able to review or score their own submissions or any submission where there may be a conflict of interest. The review process will consider the relevance, impact, and pertinence of the proposed topic(s) and also evaluate the proposed list of speakers for balance in terms of gender, career level, and the geographic spread of the represented institutions. There will be two calls (see deadlines) operating on a first-come, first-served basis between the two calls should there be slots left following the first call.

Go top of the page.

---

Deadlines

Registration (free of charge): October 15th 24th, 2025.

First Call-for-Abstracts: August 1st, 2025 (decisions sent by August 30th, 2025).

Second Call-for-Abstracts: September 1st 12th (no more extensions!), 2025 (decisions sent by September 30th October 10th, 2025).

Go top of the page.

---

Organization and Steering Committee Members

• Alfio Lazzaro
• Tiziano Müller

Go top of the page.

---

Scientific Committee Members

• Harvey Richardson, HPE UK (chair)
• Manuel Aranaz, Codee
• Ondřej Čertík, Microsoft
• Arjen Markus, Deltares Research Institute
• Austen Rainer, Queen’s University Belfast
• John Reid, JKR Associates and Rutherford Appleton Laboratory
• Brad Richardson, NASA
• Damian Rouson, LBL

Go top of the page.

---

Code of Conduct

We strive to follow the Contributor Covenant Code of Conduct to foster an inclusive and welcoming environment for everyone.

In short:

• Use welcoming and inclusive language
• Be respectful of different viewpoints and experiences
• Gracefully accept constructive criticism
• Focus on what is best for the community
• Show courtesy and respect towards other community members

Go top of the page.

---

 

Tue, November 4

9:00 AM → 9:20 AM FortranCon 2025: Welcome and Overview

Conveners: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

9:00 AM Welcome and Overview

Speakers: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

FotranCon_Welcome.pdf

9:20 AM → 10:30 AM Sessions: Tools

Convener: Alfio Lazzaro

9:20 AM Fortitude: A fast and friendly open-source linter for modern Fortran

Open-source Fortran developer tools have long lagged behind those supporting languages such as C++ and Python. Tools considered indispensable in modern development workflows – including linters and static-analysers – often have no obvious equivalent in Fortran. This gap presents challenges when onboarding new developers, hinders modernisation efforts, and negatively impacts the overall Fortran development experience.

To address these issues, we have developed Fortitude: a linter that detects bugprone code, style violations, and opportunities for modernisation. In many cases, Fortitude is able to apply fixes to these problems automatically. Originally developed as an internal tool in an academic setting, Fortitude has found a large and growing user base. By leveraging existing tools like Ruff (an open-source Python linter) and Tree-Sitter (a general-purpose language parser framework), Fortitude delivers high-performance, robust Fortran parsing, and a user interface that will be familiar to many Python developers. Reusing these proven technologies has also greatly accelerated development compared to the ‘build from scratch’ approach that has inhibited similar projects, enabling Fortitude to jump straight to the cutting-edge.

This talk will showcase Fortitude’s key features and present a roadmap for future enhancements, including improved preprocessor support and a code formatting mode.

Speaker: Liam Pattinson (University of York)

fortrancon2025_liampattinson_fortitude.pdf

9:40 AM Codee: A Tool to Enhance Correctness, Modernization, Security, Portability and Optimization in Fortran

Codee is the best-in-class static analyzer and source code formatter for Fortran, enhancing modern Fortran software development with advanced code review and testing capabilities that have only been available for C and C++ until the moment. Powered by Codee’s Deep Analysis technology, the tool helps catch bugs, enforce coding guidelines, modernize legacy code, ensure code portability, address security vulnerabilities, and optimize code efficiency of Fortran/C/C++ code. The first part of the talk introduces the Codee tool, including free-form and fixed-form Fortran source code formatting, linter-like reports, prioritization rankings, AutoFix’es, incremental static analysis capabilities and integrations with IDEs, control version systems and CI/CD frameworks. The second part of the talk presents several contributions to the Fortran community, including automated testing of the open-source projects TreeSitter, LLVM, LLVM/Flang for large, real-world Fortran codes, as well as initiatives like the Open Catalog of checkers for correctness, modernization, security, portability and optimization. Lastly, the talk aligns with recent trends in DevOps and DevSecOps by presenting Codee as the tool-of-choice to enforce compliance with secure coding standards like CWE, CERT and ISO.

Speaker: Manuel Arenaz (Codee)

FortranCon2025-Codee Slides (Last Rev Nov 3rd 2025).pdf

10:10 AM Fortuno: Flexible & extensible object oriented Fortran unit testing framework for serial, MPI-parallel and coarray-parallel projects

Fortuno is a flexible and extensible unit testing framework tailored for modern Fortran programming. It offers a simple, user-friendly interface requiring only minimal amount of boiler-plate code when writing unit tests. It puts strong emphasis on modularity and extensibility, providing a robust foundation for creating customized unit testing environments. It is written in Fortran 2018, and supports various testing scenarios, including testing of MPI- and coarray-parallel projects.

In this contribution, I'll give an overview over the architecture of Fortuno, demonstrate its usage in selected testing scenarios and discuss some best practices.

Speaker: Dr Bálint Aradi

FortranUnitTesting.pdf

10:30 AM → 10:50 AM Break

10:50 AM → 12:00 PM Sessions: Coding

Convener: Austen Rainer (Queen's University Belfast)

10:50 AM Simulating the Universe: My wish list after 20 years of Fortran development

Fortran unlocked our ability to perform realistic simulations of the Universe, including the formation and interaction of galaxies, stars and planets. In this short talk I will present lessons learnt from 20 years of developing Fortran simulation and visualisation codes for astrophysical fluid dynamics. This will consist of an unsorted list of praise, gripes and wishes for the underlying language and how we can coordinate its ongoing development to persist as the bedrock of fast, reliable astrophysical simulation codes.

Speaker: Daniel Price (Monash University)

price-fortcon-nov2025-lowres.pdf

11:10 AM Modern Fortran for Fluid–Structure Simulations of Cilia and Particles

Modern Fortran offers powerful features for building extensible and efficient solvers in computational science. We present a fully Fortran based framework for simulating coupled fluid–structure interactions involving arrays of flexible cilia and deformable particles immersed in flow. The solver employs the immersed boundary method to couple a Navier–Stokes fluid solver with Kirchhoff rod models for elastic cilia and finite element models for deformable particles.
Key design elements include the use of derived types, allocatable arrays, and modular programming to represent cilia and particles as objects with geometry, forces, and update procedures. This object-oriented approach enables handling large arrays of interacting structures with clarity and efficiency. The entire solver, including fluid, solid, and coupling modules, is implemented in Fortran, demonstrating its
continuing relevance for multiphysics simulations. GPU acceleration of the pressure Poisson equation is optionally achieved through Fortran–C interoperability with NVIDIA’s AmgX library.
This work illustrates how modern Fortran facilitates high-performance scientific computing while supporting modular, maintainable code design. The presented cilia–particle interaction study show-cases Fortran’s capability to address complex biological and physical problems in computational fluid dynamics.

Speaker: Divyaprakash Divyaprakash (Department of Applied Mechanics, I.I.T. Delhi, India)

3dparticle.mp4

FortranCon_Presentation.pdf

simulation_10.mp4

11:30 AM Modernizing OpenRadioss: A Journey Through Three Decades of Fortran Evolution

OpenRadioss (https://github.com/OpenRadioss/OpenRadioss) represents the open-source version of Radioss, a mature finite element solver with over 30 years of continuous development and about two million lines of Fortran code. The code base is predominantly written in Fortran with some C/C++ components, reflecting decades of engineering expertise in computational mechanics and crash simulation.

As a publicly available open-source code base, OpenRadioss empowers a worldwide community of researchers, software developers, and industry leaders to tackle the latest challenges in rapidly evolving technologies. These include battery development, lightweight materials and composites, human body models and biomaterials, virtual safety testing.

Currently, an ambitious modernization effort is underway to transform this legacy code base by adopting more recent Fortran standards and implementing contemporary best practices. This initiative aims to enhance code maintainability, performance, and developer productivity while preserving the solver's proven reliability and accuracy.

This presentation will share our developer experience and lessons learned. We will discuss the benefits of adopting modern Fortran features, and the practical challenges encountered along the way. Beyond language features, we will also explore the broader ecosystem of tools essential to this modernization effort, including compilers, static analyzers, build systems, and other development tools.

Speaker: Mr Olivier Wienholtz

FortranCon_OlivierWienholtz_November_4th_2025.pdf

12:00 PM → 12:10 PM FortranCon 2025: Close Morning Session Day 1

Conveners: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

12:00 PM Close Morning Session Day 1

Speakers: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

4:40 PM → 5:00 PM FortranCon 2025: Opening Evening Session Day 1

Conveners: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

4:40 PM Opening Evening Session Day 1

Speakers: Tiziano Müller (FortranCon Association), Alfio Lazzaro

5:00 PM → 6:00 PM Keynote: John Reid

Convener: Dr Harvey Richardson (HPE)

5:00 PM Keynote: Reflections on working for 45 years on the Fortran standard

I began getting seriously interested in the Fortran Standard in the late
1970s because my group used it for the sake of portability and I wanted
improvements. Since then I have worked on a wide variety of features.

My first meeting of WG5 (then known as "Fortran Experts Group") was in
October 1980 and of J3 (then X3J3) in January 1981. I was a member of
J3 from 1983-1991 and its secretary 1987-1990. I was there when it got
into deadlock, solved by WG5 at the 1988 Paris meeting.

After 1991 I switched to WG5, which meets only annually and I was
convener from 2000-2017. I still attended J3 meetings as an alternate
member when there was a special need and now attend only by zoom.

My plan for this talk is to tell you both about the technical issues
that I supported strongly and about the problems over reaching
consensus that I faced in J3 in the 1980s and as convener during my
long role. I hope you will find this interesting and I hope to convince
you of the merits of using a committee to form a standard.

Speaker: Dr John Reid (Rutherford Appleton Lab. UK)

FortranCon25.pdf

6:00 PM → 7:00 PM Sessions: Fortran and Python

Convener: Dr Harvey Richardson (HPE)

6:00 PM Gfort2py: A simple way to call Fortran from Python

Here I will discuss my open source project Gfort2py, a library designed to make it simple to call Fortran from Python, which has been developed over the last 10 years. Gfort2py was designed to make it simple to interface Fortran to Python without needing to make any changes to the Fortran code, while also minimizing changes to build systems. I will discuss how you can interface Fortran with Python, without having to rewrite your code to use iso_c_binding or otherwise annotating your source code to make Python understand Fortran. I will show some of the problems I encountered along the way in understanding Fortran code and creating an interface to it. I will demonstrate its abilities to call Fortran procedures and access module variables, this will also include demonstrating some of its advanced capabilities including accessing allocatable arrays, derived types, and IEE 754 quadruple precision variables.

Speaker: Robert Farmer

gfort2py_2025.pdf

6:30 PM Interfacing Fortran and Python for ODE Modeling: A Performance Study on Ethylene Dichloride Cracking

This presentation explores the integration of Fortran and Python in solving a system of ordinary differential equations (ODEs) modeling the thermal cracking of Ethylene Dichloride (EDC), a process relevant to vinyl chloride production. The reaction system consists of two consecutive first-order reactions and is solved using the Runge-Kutta-Gill method.

The Fortran implementation is compiled and exposed to Python using f2py, enabling interoperability with Python’s visualization and benchmarking ecosystem. A pure Python implementation using the same numerical method is developed for comparison. Performance benchmarks reveal that the Fortran-based solver achieves speedups of up to ~700× compared to the Python implementation, particularly at higher step counts.

This study demonstrates the complementary strengths of Fortran and Python: Fortran’s numerical efficiency and Python’s ease of prototyping and analysis. All source code, including Fortran modules, Python wrappers, and benchmarking scripts, is made available as part of an open-source repository. The results underscore the value of mixed-language workflows for modern scientific computing, especially in domains requiring both high performance and rapid experimentation.

Speakers: Helianildes Silva Ferreira (Federal University of Bahia), Rodrigo Silva Ferreira (Posit PBC)

FortranCon 2025 - Quarto doc (follow along)

FortranCon 2025 - Rodrigo and Helianildes.pptx

GitHub Repository

LinkedIn - Helianildes Silva Ferreira

LinkedIn - Rodrigo Silva Ferreira

7:00 PM → 7:20 PM Break

7:20 PM → 8:50 PM Sessions: Coding

Convener: Brad Richardson (Amentum)

7:20 PM Precision-Agnostic BLAS/LAPACK and a Modern Linear Algebra API for the Fortran Standard Library

Modern scientific workloads demand linear-algebra kernels that are both performant and expressive. We present the new precision-agnostic BLAS and LAPACK layer being upstreamed to the Fortran Standard Library (stdlib). The reference codes have been fully modernised and templated, delivering real and complex arithmetic in 32-, 64-, 80- and 128-bit kinds and supporting the 64-bit integer interface to scale past two-billion-element arrays and work with vendor-optimised back-ends. With one macro, 32- and 64-bit real procedures can be transparently redirected to an optimised third-party library (OpenBLAS, MKL, Accelerate, …), while stdlib continues to provide full kind-agnostic coverage.

Built on these kernels, we provide a NumPy/SciPy-style API that retains Fortran’s zero-overhead semantics: pure functions, allocation-free subroutines and intuitive operators for determinants, inverses, factorisations and other advanced solvers. A lightweight state handler offers optional, zero-cost error handling suitable for both scripting-style and HPC codes.

The talk will cover design challenges, templating strategy, one-macro integration and head-to-head performance against Python front-ends, and highlights current community contributions and roadmap items. Attendees will learn how to adopt, extend and benchmark this next-generation Fortran linear-algebra stack, positioning stdlib as an everyday alternative to established numerical platforms.

Speaker: Federico Perini (Wisconsin Engine Research Consultants, LLC)

FedericoPerini_PrecisionAgnostic_BLAS_LAPACK.pptx

GitHub materials and presentation

7:50 PM Using Modern Fortran OOP for the development of an open-source extensible library for thermodynamic calculations

Thermodynamic calculations are fundamental in chemical engineering, with applications in CFD, process and reservoir simulations, among others. Most of
these calculations are based on Equations of State and their respective derivatives that allow constructing the thermodynamic properties required. Equations
of State (EoS), expressed either for Pressure or residual Helmhholtz energy, are
functions dependant on always the same variables (⃗n, V, T ). So the algorithms
used to calculate properties can be written in an abstract manner, independent
of the specific EoS.
We present yaeos, and how an OOP approach has helped making modular code.
It is easy to extend with new models and algorithms internally and when used as
a dependency. It also implements an extra module that uses ISO C BINDING to
work as a C interface. This interface is also used to compile a Python extension,
using f2py.
To generate the Python extension meson is used as a build system. To
build the library it runs fpm internally and then f2py to build the C interface
and extra Python files. The library includes continuous integration testing for
both interfaces, and automatically uploading Python wheels. Documentation is
generated using FORD and Sphinx. Documentation and source code available
at: https://github.com/ipqa-research/yaeos

Speaker: Federico Ezequiel Benelli (IPQA (UNC-CONICET))

FortranCon-2025.pptx

GitHub Repository

Our institutional repository

8:20 PM Experiences with Idiomatic Vibe Coding

This talk will report on experiences with using recently developed correctness-checking idioms in prompt engineering. The idioms read as natural language statements, a property they share with "vibe coding". At least two properties distinguish the idioms from other vibe coding. First, the idioms are backed up the Fortran standard's formal grammar as defined in extended Backus-Naur form. Second, the resulting statements have unambiguous definitions in the form of defined operations from by the Julienne correctness-checking framework. As a baseline for comparison, we first used a free-form English to prompt a large language model (LLM) to generate a finite-difference approximation to a three-dimensional Laplacian operator for the Matcha T-Cell motility simulator's diffusion solver. Inspired by agile software development's test-driven development approach, we then refined our prompt by introducing Julienne tests for validation. We found that even for writing one function that junior research software engineers could likely write without difficulty, three of four LLMs prompted with precise terminology from the Fortran standard produced code with compile-time errors after two iterations. One model generated compilable code with vibe coding. Idiomatic vibe coding reduced the number of iterations required to generate compilable code but also introduced a regression: the test failure count increased.

Speaker: Dr Damian Rouson (Berkeley Lab)

Experiences with Idiomatic Vibe Coding.pdf

8:50 PM → 9:00 PM FortranCon 2025: Close Evening Session Day 1

Conveners: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

8:50 PM Close Evening Session Day 1

Speakers: Tiziano Müller (FortranCon Association), Alfio Lazzaro

Wed, November 5

9:00 AM → 9:10 AM FortranCon 2025: Opening Morning Session Day 2

Conveners: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

9:00 AM Opening Morning Session Day 2

Speakers: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

9:10 AM → 10:20 AM Sessions: Coding

Convener: Arjen Markus (Deltares)

9:10 AM From Fortran to Fortran: The Porting Guide

Porting Fortran applications across different compilers often exposes hidden assumptions, subtle standard deviations, and possible bugs. Modern compilers provide a rich set of diagnostics and runtime checks that go far beyond basic compilation, offering insights into code structure, correctness, and maintainability. This talk presents strategies for using compiler warnings, stricter checks, and optimization feedback to identify structural weaknesses, guide refactoring, remove non-standard constructs, and detect hidden bugs. Attendees will learn practical techniques for turning compiler feedback into improvements, creating reliable and maintainable Fortran applications.

Speaker: Igor Gerasimov (Budapest University of Technology and Economics)

presentation.pdf

9:40 AM Practical performance benefits of the Fortran PURE keyword in a CFD solver

The PURE keyword, introduced in Fortran 95, imposes constraints on a procedure – namely that it cannot perform side effects such as I/O (e.g. printing to the screen). PURE procedures have benefits from a software engineering perspective, making programs easier to test and guiding their development – as shown in https://pbartholomew08.github.io/thoughts/testing_pure_fortran.html. Another potential benefit is that by providing the compiler with additional information, PURE may enable additional optimisations. During the ASiMoV project, we undertook an effort to increase the use of PURE in the ASiMoV-CCS CFD solver, with the goal of improving code quality based on our experiences with using PURE in Fortran elsewhere. As most of the procedures this was applied to were relatively small we did not expect to see significant performance benefits, however with minimal changes we obtained significant improvements in several key components of the solver and an approximately 20% performance benefit overall. In this presentation we will assess the causes of these benefits and whether it is compiler dependent.

Speaker: Dr Paul Bartholomew (EPCC, University of Edinburgh)

PURE FORTRAN.pptx

10:00 AM State of the FOSS (Fortran Open Source Software)

Fortran has a distinguished history as one of the earliest programming languages, yet its modern open source ecosystem is often perceived as small or stagnant compared to more recent languages. In this talk we present an evidence-based view of the current state of Fortran open source software by mining activity across GitHub. Our study examines the top 2000 “most relevant” Fortran repositories and compares them with equivalent sets in Python, Java, Go, and Node. We consider both absolute measures and values normalised against estimated user bases, focusing on active contributors, commit frequency, issue activity, and repository health.

The analysis reveals patterns that distinguish Fortran projects: concentrations in numerical libraries, high-performance computing, and domain-specific codes; lower churn but relatively stable maintenance; and a smaller but committed contributor base. When adjusted for community size, Fortran activity compares more favourably than raw numbers suggest, particularly in terms of longevity and resilience of core projects. By highlighting both strengths and weaknesses, this work provides a clearer picture of how Fortran fits into the wider open source landscape today. We conclude with implications for sustainability and growth, and discuss opportunities to strengthen Fortran’s role in the modern open source ecosystem.

Speaker: David Cutting (Queen's University Belfast)

DC_FOSS.pdf

DC_FOSS.pptx

10:20 AM → 10:40 AM Break

10:40 AM → 12:00 PM Sessions: Compilers

Convener: Manuel Arenaz (Codee)

10:40 AM Fortran, the OpenMP API, and AMD GPUs - Porting Applications for Performance

Today’s supercomputers are massively parallel, heterogeneous systems and GPUs have long been established as accelerators to boost the execution of data parallel algorithms. However, programming GPUs remains complex and requires specialized expertise to optimize code for a specific architecture, oftentimes involving a vendor-specific programming model. This usually results in code that is not portable between GPU vendors, typically requiring that programmers have to maintain multiple versions to target different architectures. Many software packages reflect this circumstance and provide tailored implementations for HIP, CUDA, and OpenCL. This talk has two parts. Firstly, it will introduce target offloading using the OpenMP API, an easy and portable way to exploit a GPU's parallelism. We will review some basic OpenMP directives and allude to key concepts of the heterogeneous features of the OpenMP API. Secondly, the talk will highlight some recent case studies of (large) Fortran codes that have been ported to AMD Instinct™ Accelerators using the AMD Next-Generation Fortran Compiler, based on LLVM/Flang. For the case studies, we will sketch the way the OpenMP API was applied and review the performance obtained on AMD Instinct™ Accelerators.

Speaker: Michael Klemm (AMD)

8183 2025-11-05 FortranCon 2025 A+A+A.pdf

11:10 AM Status of gfortran, the gcc fortran compiler

The purpose of the GNU Fortran (GFortran) project is to develop the Fortran compiler front end and run-time libraries for GCC, the GNU Compiler Collection. The GFortran compiler is fully compliant with the Fortran 95 Standard and includes legacy F77 support. In addition, most F2003 and F2008 and an increasing number of F2018 and F2023 features are implemented. Of particular importance are the support for MP and coarrays to support production parallel computation. See the gfortran wiki for more details: https://gcc.gnu.org/wiki/Gfortran.

The presentation will describe the status of implementation of the new or modified features of each of the newer standards and the areas where active development is concentrated.

GFortran development follows the open development process. We do this to attract a diverse team of developers and to ensure that GFortran works on multiple architectures and diverse environments. We always need more help! Please visit https://gcc.gnu.org/fortran/ and the links therein.

1] https://gcc.gnu.org/onlinedocs/gcc/Contributors.html identifies all the gfortran contributors. Those that are currently active can be identified in recent posts to the gfortran mailing list: https://gcc.gnu.org/ml/fortran/

Speaker: Paul Thomas

FortCon2025.pdf

11:30 AM Fortran - What are the options for accelerated computing?

This case study summarizes the evolution of an existing CUDA code base in DBCSR and CP2K like the introduction of HIP as well as an OpenCL based implementation. The offload interface in CP2K for instance is already an evolution of DBCSR's offload interface with both focusing on commonly supported primitives accross stream programming models, i.e., only a handful of easy-to-interface C functions.

With the introduction of HIP, the idea extended to limiting kernel language to a well supported (sub-)set of primitives. For the OpenCL side, the work raised the limits of the existing implementation in DBCSR such as improved auto-tuning infrastructure, generalized kernels and tuning parameters, supporting all vendors out of the box, and reusing the OpenCL backend with CP2K Molecular Dynamics application.

The talk closes with a collection of results achieved on current HPC installations (DBCSR-MM and CP2K-DBM distributed block-sparse matrix multiplication).

Speaker: Hans Pabst (Intel)

FortranCon25.pdf

12:00 PM → 12:10 PM FortranCon 2025: Close Morning Session Day 2

Conveners: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

12:00 PM Close Morning Session Day 2

Speakers: Tiziano Müller (FortranCon Association), Alfio Lazzaro

4:50 PM → 5:00 PM FortranCon 2025: Opening Evening Session Day 2

Conveners: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

4:50 PM Opening Evening Session Day 2

Speakers: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

5:00 PM → 5:30 PM Invited talk

Convener: Tiziano Müller (FortranCon Association)

5:00 PM Invited talk: Fortran Standardization Process and the Road to F202Y

In this talk, I will start with an overview of the Fortran standardization process, including how the standard committee work and how to contribute to the process. I will then discuss the current status of the Fortran standard, including features being considered and developed for the next release, currently designated as F202Y. Finally, I will provide a brief preview of some exciting features that are planned for inclusion in F202Y.

Speaker: Reuben Budiardja (Oak Ridge National Laboratory)

2025-11-FortranCon_FortranStandard_and_F202Y-1.pdf

5:30 PM → 6:30 PM Sessions: Coding

Convener: Tiziano Müller (FortranCon Association)

5:30 PM Compile-time unit checking in Fortran with genunits

Automatic checking of the units in Fortran programs has long been proposed to catch bugs. Implementations have often been limited to test problems and not been applied in practice. Consequently, there has not been much practical experience with unit checking in Fortran. In this work, a compile-time unit checking approach has been developed with gennuits. genunits will generate a Fortran module with one derived type per unit. Valid operations between derived types/units are defined, leaving invalid operations undefined. Consequently, a units error will result in a compilation error. Some significant practical limitations of this approach to unit checking have been identified. Run-time performance can be significantly slower without inlining. Compilation can become far slower or even impossible, and genunits has been designed specifically to help mitigate this problem by generating only necessary derived types. Compiler error messages also can be unclear as they were not written with this application in mind. It is hoped that the identified issues can be mitigated in future Fortran language and compiler developments.

Speaker: Ben Trettel

genunits.pdf

genunits.zip

5:50 PM (Let's) make Fortran spatial again: Re-establishing Fortran as a language for spatial analysis

Fortran's dominance in high-performance computing (e.g., climate modeling, finite element analysis, reactive transport in porous media) is hindered by its lack of native support for geospatial data operations. Fortran Numerical libraries often rely on cumbersome custom I/O routines for reading, writing, and reprojecting spatial data. This work presents a solution: the development and implementation of Fortran bindings with modern interfaces to the foundational geospatial libraries PROJ (library for coordinate transformation) and GDAL (library to read geospatial data) Their practical application is demonstrated through integration into ATALIB, a geostatistical library implemented in modern Fortran. This integration decouples ATALIB's computational core from its I/O limitations, enabling direct access to data from cloud storage, web services, and hundreds of formats via GDAL, with precise spatial coordinate and transformation ensured by PROJ. A case study on a large 2D kriging operation benchmarks the new implementation against a legacy version, quantifying significant gains in interoperability and processing efficiency. The development of Fortran bindings to the geospatial libraries GDAL and PROJ is a transformative strategy that allows the Fortran community to improve the interoperability of decades of performance optimized code without compromising computational efficiency.

Speaker: Prof. Oscar Garcia-Cabrejo (Universidad Pedagógica y Tecnológica de Colombia)

FortranCon2025_OscarGarciaCabrejo.pdf

6:10 PM Parallel Monte Carlo and Machine Learning Tools for Fortran

ParaMonte (short for Parallel Monte Carlo: https://github.com/cdslaborg/paramonte) is a library of Monte Carlo and Machine Learning routines, available in both serial and parallel versions, for general common scientific inference, particularly tasks involving data analysis, optimization, sampling, and integration of mathematical density functions, and exploration of the posterior distributions in Bayesian models and inverse problems. The library has been designed with a strong focus on unifying automation, ease of use, high performance, scalability, and reproducibility. We will discuss the library's facilities for end-users and its evolution to a multi-precision, fully-generic interface library, comprising nearly 2000 functionalities for machine learning and scientific inference.

Speaker: Prof. Amir Shahmoradi (University of Texas)

main.pdf

6:30 PM → 6:50 PM Break

6:50 PM → 7:50 PM Sessions: Compilers

Convener: Dr John Reid (Rutherford Appleton Lab. UK)

6:50 PM LFortran - Modern interactive LLVM-based Fortran compiler

LFortran is a modern, open-source interactive Fortran compiler designed for high-performance computing and scientific applications. Built on top of the LLVM infrastructure, it offers a robust, modular architecture that facilitates advanced code analysis, optimization, and portability. While LLVM remains the primary backend in an advanced alpha stage, LFortran also supports experimental backends such as C, C++, Julia, MLIR, and WebAssembly, enabling cross-language interoperability and broader platform support. Recent developments have significantly enhanced LFortran’s capabilities, including successful compilation of complex scientific projects like PRIMA, POT3D, and SNAP, as well as support for OpenMP pragmas, the concurrent keyword, and compile-time evaluation of arrays. To ensure supporting legacy features, LFortran compiles 60% of Fortran code leveraged in SciPy. The compiler is also making strides in GPU offloading for NVIDIA architectures and web-based execution. LFortran provides partial support for Fortran Package Manager (FPM) and the growing Fortran standard library (stdlib). Its object-oriented features are being actively developed to better support modern Fortran codebases. Its ongoing improvements aim to make it a full-featured, production-ready compiler that meets the demands of modern Fortran developers while integrating seamlessly into heterogeneous computing environments.

Speakers: Ondřej Čertík (Microsoft), Pranav Goswami

LFortran-FortranCON-2025.pdf

7:20 PM The Intel® Fortran Compiler Status Update November 2025

This session provides a status update on the Intel® Fortran Compiler, (ifx). This session provides a timeline for the 8-year development cycle for ifx and the front-end integration into the LLVM framework. We will give a status update on Fortran 2023 features along with OpenMP 6.x features. In addition, a preview of features coming in 2026. Learn about the development journey and get a better understanding of ifx features today. Time will be provided for questions and feedback on ifx.

Speaker: Mr Ronald Green (Intel Corporation)

Intel_Fortran_Update_2025November.pdf

7:50 PM → 8:30 PM FortranCon 2025: Final thoughts / Discussions / Closeout

Conveners: Alfio Lazzaro, Tiziano Müller (FortranCon Association)

7:50 PM Final thoughts / Discussions / Closeout

Speakers: Tiziano Müller (FortranCon Association), Alfio Lazzaro

FotranCon_Closeout.pdf