This Coding Café session was a practical introduction to cloud computing for research, and to SURF Research Cloud in particular.
Topics covered:
- What "cloud computing" means in a research context, and when it is actually worth using
- Signs you have outgrown your laptop: running out of memory or storage, analyses that take hours or days, software that is awkward to install locally
- Trade-offs to weigh: upload and download times for large data, costs accruing while an instance runs, less institutional IT support, and the risk of leaving instances running
- The landscape of options — public clouds, HPC and cluster computing, platform services, and what is available at TU/e
- Choosing resources sensibly: how many vCPUs, how much RAM, and why extra cores do not help work that is not parallel
- Logging in over SSH with key-pair authentication, followed by a live SURF Research Cloud workspace demo
A key takeaway from the session: you pay for running resources, not for finished jobs — so always stop or shut down your workspace when you are done.
