AI Thoughts on AI for 2025 I am in the fun position not only of using AI as engineer but also need to making hiring decisions in the face of tools such as Cursor and Devin. I thought I'd share some thoughts.
Random Bytes Dreams and Wishes for HPC in 2025 A contrarian look at what I hope will be forthcoming for HPC in 2025 rather than any real predictions.
HPC HPC on Cloud — Cost Optimisation & Cloud Economics A look at the economics of running a HPC system on a public cloud infrastructure. It looks at the impact of the performance differential between different types of VM, the impact of price variation and introduces the concept of a performance normalised price to allow easier comparison of VMs.
HPC HPC on Cloud for Financial Risk — A Comparative Analysis This article presents the results of running the COREx benchmark on over 650 different VM types across the three major cloud providers. It attempts to contextualize these results and help HPC practitioners interpret them to optimise their systems, as well as discuss the methodology employed.
HPC Hyperthreading in HPC: On or Off? This article discusses the use of simultaneous multi-threading (hyper-threading) in HPC, with a particular emphasis on financial risk systems and the COREx benchmark. It looks into the SMT status of cloud virtual machines, the ability to enable or disable and the cost implications of doing so.
HPC COREx - A Financial Risk HPC Benchmark Evaluating hardware performance such that it correlates well to your own use case is always difficult.Financial risk system specific benchmarks can be even more of a challenge due the proprietary and often secret nature of the code involved. This article looks at developing such a benchmark.
HPC HPC Glossary A glossary of (HPC) terms as used in our recent articles to ensure clarity and a shared understanding. These are largely well understood; however, there can sometimes be a little confusion, especially around the common usage of the words core and CPU.
HPC High Performance Computing on the Cloud Running HPC clusters is much easier on the cloud where there’s unlimited capacity and no scheduling problems right? That’s certainly what all the cloud vendors always tell me, but how true is it?