Development of benchmarks in excalibur and measuring performance on HPC systems.

11 Jul 2025 - Luthyano F - Learning, ARCHER2, Parallel Computing, Computer Simulations

---

Introduction _______________________________________________

This week, I recapped the introduction to High Performance Computing (HPC) through EPCC and revisited what defines a supercomputer system, such as ARCHER2 and ARCHIE. I explored examples and explanations within HPC to better understand how to run benchmarks and communicate jobs from a laptop to an HPC system like the one at UCL. This content has deepened my understanding of the purpose of high performance computers, the importance of parallel computing, and CPU architecture.

Additionally, I learned more about the structure that makes a laptop functional, which has improved my understanding of computer hardware and its impact on overall system performance.

---

Challenges _______________________________________________

I faced some challenges while revisiting the Introduction to HPC, as I needed to relearn key concepts such as what enables parallel computing and why simply adding more CPUs does not always lead to better performance. I reviewed test examples to gain a better understanding of the types of tests that can be performed on a supercomputer or HPC system.

Finally, I explored the theoretical idea of recreating ARCHIE using five Arduinos, as suggested in weeks 2–3 of the Introduction to HPC.

---

Stress Relief _______________________________________________

Public domain from Freepik