02_environment.md

---

theme: csc-eurocc-2019
lang: en

---

Brief introduction to HPC environments {.title}


All materials (c) 2020-2025 by CSC – IT Center for Science Ltd. This work is licensed under a **Creative Commons Attribution-ShareAlike** 4.0 Unported License, [http://creativecommons.org/licenses/by-sa/4.0/](http://creativecommons.org/licenses/by-sa/4.0/)

Notes on vocabulary

- You can roughly think that one **node** is a single computer - A node on a supercomputer contains: - One or more central processing units (**CPUs**) with many **cores** - Shared **memory** - Some nodes may also have: - **Local storage** (NVMe) - Graphics processing units (**GPUs**)

{width=60%}

Cluster systems

- Login nodes are used to set up jobs (and to launch them) - Jobs are run on the compute nodes - A batch job system (scheduler) is used to run and manage the jobs - On CSC machines, we use **Slurm** - Note: Slurm uses term CPU for cores, which is a bit confusing

{width=80%}

Available HPC and cloud resources

• Puhti is the general-purpose supercomputer ☑️

• Mahti is the massively parallel flagship supercomputer

• LUMI is a European pre-exascale supercomputer operated by CSC

• Pouta provides cloud resources via OpenStack (IaaS)

• Rahti provides containers via OKD (PaaS)

• Allas provides object storage for all services

Coming early 2026:

• Roihu is CSC's next national supercomputer, will replace Puhti and Mahti

Which supercomputer to use?

• What kind of resources can your application use?

  • Can it use more than one core?

  • How much memory will it need?

  • Can it use a GPU or an NVMe?

  • What takes long (i.e., the time-limiting part) in your job?

• See what kind of resources are available?

  • Is my code already installed?

  • Max runtime, partitions (queues), provisioning policy (Per core/per node/other)

  • Each system is different, so check the documentation

Quick and dirty comparison of Puhti, Mahti and LUMI

	Puhti	Mahti	LUMI
Pre-installed apps	120+	40+	See here
Cores per node	40	128	128
Job size (min-max cores)	1-1040	1-25600	1-65536
Memory per node (GiB)	192-1536	256	256-1024
GPU cards	320 (V100)	96 (A100)	11912 (MI250X)
Nodes with NVMe (CPU+GPU)	106+80	60+24	8+8

In short: Mahti is for large parallel jobs, prepare to install and optimize your code. Still, one Puhti *node* is 10x your laptop. LUMI is like Mahti + massive AMD GPU capacity