Large Language Model Successfully Ports Complex Ocean Simulation Model to Modern Computing Languages
Researchers used an agentic large language model (LLM) coding assistant to translate the FESOM2 production ocean-sea-ice model from approximately 74,000 lines of Fortran into C and then C++/Kokkos, preserving the model's physics throughout. The project required domain-expert oversight and a strict two-stage translation process that separated numerical reproduction from parallelism introduction. The result demonstrates that LLMs can port complete, production-scale scientific codes to modern GPU-accelerated frameworks in a matter of weeks, reaching the performance thresholds needed for operational climate simulations.
A team of researchers has demonstrated that an LLM-assisted agentic coding assistant, guided by domain experts, can successfully port FESOM2—a full production unstructured-mesh ocean and sea-ice model—from roughly 74,000 lines of Fortran first to C and then to C++/Kokkos for performance portability across CPUs and GPUs. Three practices were identified as critical to success: a two-stage translation that first reproduced the numerics faithfully in C before introducing GPU parallelism via Kokkos; a strictly literal translation policy that prevented the LLM from autonomously 'improving' the source code; and rigorous stage-appropriate validation at each step. The C port reproduces the original Fortran results at the level of long-term simulation statistics over five-year runs, while the Kokkos port is bit-for-bit identical to the C reference on CPU and statistically consistent on GPU over multi-year integrations. On eddy-rich meshes with up to 7.4 million surface vertices, a single NVIDIA A100 GPU node runs 1.6 to 3.7 times faster than a comparable CPU node, meeting the 1–2 simulated years per day threshold required for production climate integrations. The authors document both successful practices and failure modes encountered, offering a methodological template for porting other legacy scientific codes. The work suggests that LLMs, when carefully directed, can dramatically accelerate the modernization of large geophysical software codebases that would otherwise require years of manual effort.
What's missing
The study does not report total computational cost or human expert hours required for the porting process, which would be important for assessing practical scalability to other models. The paper has not yet undergone formal peer review, as it is a preprint.
What different sources said
- arXiv physicsCenter
An Ocean Model Ported by a Large Language Model: Experience and Lessons from FESOM2 (Fortran to C to C++/Kokkos)
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.