---
tags: Kickstart
---

# Me and HPC

## Main messages
- Don't feel like a dummy
- computing is cool and useful
- computing is a profession. you may be more or less involved in doing it (or work with others)

## Plot

- xx:50 Icebreaker: how did you start your computing? Or how are you starting?
- xx:00 Begin, introduce ourselves and first talk
- Samantha: Geoinformatics, part time PhD student, researcher, and recently started as geoinformatics specialist at CSC (which few years ago was unthinkable, when starting out with scientific computing), enjoys learning new stuff about scientific computing by community of like-minded and thinking how to lower the barrier into using HPC systems
- xx:02, S/R answer the icebreaker
- Samantha:
- from this data we want to have these results
- lots of data -> not feasible to run on laptop -> CSCs HPC.
- before: commercial GUI software at university; not available for the job -> overwhelmed.
- made lots of mistakes, inefficient workflows which taught me a lot
- learning more and how to do things better every day from Aalto, CR and CSC people
- Richard: I didn't realize when I was going. I definitely didn't do things well back then, but I read a lot. was mostly alone, though.
- ???:
- xx:07 computers for science vs computational science vs scientific computing
- xx:09 What's your path?
- I need it but it's not my job
- I need it but it's just a tool
- I am focused on the tools
- (different paths for different domains)
- xx:12 Feeling of being overwhelmed
- Many people will tell you the "right" way to do it
- Freedom to do it wrong - that's how you learn
- don't confuse experience with skill
- don't let perfect get in the way of better.
- good enough practices: https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005510
- https://scicomp.aalto.fi/scicomp/zen-of-scicomp/ - "try to be better, don't worry about being perfect right now."
- https://en.wikipedia.org/wiki/Worse_is_better
- you usually can't break things that badly (but don't lose your data!).
- "Just try it. If it breaks, then you figure out how to fix it and you learned more". (others can help)
- xx:16 Community and help
- you need mentors, courses are not enough
- Should you ask for help? yes! Radovan will say more
- The first line of support is your peers. Community aspect (remote work vs in-person work).
- Group leaders: make sure people aren't so overwhelmed that they can't support each other.
- Desk next to each other and share screen. or screenshare remote.
- Ask early, it's OK
- xx:20 What's the next steps?
- rest of the course is a starting point
- learn to learn (how to read docs, web search, etc.)
- Fundamental skills that may be scary
- command line environment - connects everything together
- version control - keeps your stuff organized
- data management (micro and macro) - because it's hardly ever about the computing itself
- xx:24 What comes the rest of the day
- xx:25 End

## Notes

- fear of command line
- don't feel like a dummy, everyone once started without knowing anything about it
- the "gurus" are not gurus in what you are doing!
- experiments are important, play around!
- slowly build up, e.g. test yourself, interactive jobs. You don't need to go straight to the end
- Where is the best place to play around with basic Linux commands?
- Being able to search for things when you need it! And not be afraid of reading documentation. Types of documentation.
- hierarchical support
- Community aspect (especially remote work time, but in-office isn't necessarily better!)
- Don't be embarrassed of showing people what you are doing (and learning from other)
- getting help from staff
- go early (even if it's just for initial pointers)
- open source vs not

- never a waste of time to learn theses skills, they are reusable in other fields