Skip to content

Fluid Numerics Journal

Back to Fluid Numerics

Port, Optimize, and Scale on AMD Instinct GPUs with HotAisle

Hot Aisle Servers in Grand Rapids, Michigan With unmatched performance for HPC and AI workloads, scaling on AMD MI300X hardware can transform your computational capabilities—but only if your software is optimized to take full advantage. Discover how Fluid Numerics and Hot Aisle can help you seamlessly port, optimize, and scale your applications to achieve breakthrough results.

Accelerating Science with GPU Software Optimization

Perfetto trace profile example What does it take to become a finalist for the prestigious Gordon Bell Prize?

For the EXESS team, it started with a focused effort to optimize their quantum chemistry application for cutting-edge GPU hardware. Discover how our Mentored Sprint service helped them achieve breakthrough performance, unlock new possibilities, and earn recognition on the world stage. Read more

Maximizing Performance, Minimizing Costs: Energy Savings from GPU Optimization

Final performance tables In high-performance computing, optimizing GPU workloads isn’t just about speed—it’s about unlocking hidden savings in energy and sustainability. Discover how a 1.91x performance boost turned into real cost savings and why software optimization could transform your operations. Read more

What is a mentored sprint ?

Mentored sprint thumbnail Imagine achieving months of software optimization progress in just one week.

Whether you're optimizing performance, porting to new hardware, or tackling costly inefficiencies, our Mentored Sprint service delivers fast, measurable results. Discover how teams are transforming their applications, cutting costs, and future-proofing their software with expert guidance. Read more

HIP Performance Comparisons : AMD and Nvidia GPUs

Spectral Element Mesh If you've read some of my other posts, you're aware I'm in the midst of refactoring and updating/upgrade SELF-Fluids. On the upgrade list, I'm planning a swap-out of the CUDA-Fortran implementation for HIP-Fortran, which will allow SELF-Fluids to run on both AMD and Nvidia GPU platforms. This journal entry details a portion of the work I've been doing to understand how some of the core routines in SELF-Fluids will perform across GPU platforms with HIP. Read more