Projects on Fortran 2026

Profitool

Mon, 01 Jan 0001 00:00:00 +0000

We are experts on Profitool, a long-running Fortran ERP. We have migrated a production Profitool deployment from 32-bit to 64-bit, added modern language features, and built the testing and deployment pipeline around it. If you run Profitool — or a similar Fortran-based ERP — we can take it forward instead of forcing you to start over.

Where we add value

32-bit to 64-bit migration without losing decades of business logic — see Expertise → 32 → 64-bit migration.
Modern extensions: SQL database access, regular expressions, hash maps, and REST APIs added directly into the existing Fortran codebase.
Containers and reproducible builds so the same binary you test is the binary you ship.
A real test framework — TAP-style unit tests with prove runners, instead of “compile and pray”. See Examples → Unit Tests.
Day-to-day operations: backup, restore, data recovery, and version-control discipline around a codebase that may have started before any of those existed.

Talk to us

simon@unisolve.com.au or scottp@dd.com.au

WII

Mon, 01 Jan 0001 00:00:00 +0000

WII is a wave-maths code written in Fortran. It runs the numerical modelling that sits behind sea-state and oceanographic products — the kind of long-running scientific Fortran where every line has been tuned over years and “just rewrite it” is not an option.

Where we add value

Maintaining and extending the wave-maths core without breaking the numerical results that downstream systems depend on.
Capturing reference outputs as regression tests so any change to the maths is provably bounded — see Expertise → Testing.
Modernising the build and runtime: pinned compilers, dependency stacks, container packaging — see Expertise → Packaging.
Recovering and migrating historic WII runs and inputs from old storage — see Expertise → Legacy data.
64-bit upgrades of older WII deployments without losing fidelity.

Talk to us

simon@unisolve.com.au or scottp@dd.com.au

WRF

Mon, 01 Jan 0001 00:00:00 +0000

The Weather Research and Forecasting (WRF) model is a large, mature, Fortran-heavy scientific code that is famously fiddly to build and operate at production scale. We help teams running WRF treat it as a real production system rather than a research experiment.

Where we add value

Reproducible builds of WRF and its dependencies (HDF5, NetCDF, MPI, compiler stacks) — pinned, scripted, containerised.
Operational packaging so a forecast run can be reproduced months later from the same inputs and binaries — see Expertise → Packaging.
Pipeline integration: feeding WRF outputs into downstream systems, dashboards, and APIs.
Recovery of legacy WRF runs and inputs from old storage — see Expertise → Legacy data.
Performance and 64-bit upgrades for older WRF deployments.

Talk to us

simon@unisolve.com.au or scottp@dd.com.au