Overview

Objectives

• We'll get an overview of UPPMAX and SNIC/NAISS and how a computer cluster works

UPPMAX = UppMACS - Uppsala Multidisciplinary Center for Advanced Computational Science

NAISS

• National Academic Infrastructure for Supercomputing in Sweden

• Mission: to provide a quality high-performance computing environment nationally

• Starting 1 January 2023, the National Academic Infrastructure for Supercomputing in Sweden (NAISS) is the new organization for high-performance computing, storage, and data services for academic users in Sweden.

• From the users' perspective, there will initially only be minimal differences between the SNIC and NAISS regimes.
  • FAQ:s — https://www.naiss.se/
• Application rounds: https://www.naiss.se//#application-rounds-for-compute-and-storage-resources
• NAISS and Uppsala University fund UPPMAX — UU's supercomputing center.

UPPMAX missions

• Runs the clusters placed in Uppsala.
• More details in the afternoon about Organisational orienteering!

UPPMAX systems

• Clusters
  • Rackham (general purpose)
    • Snowy (Long runs and GPU:s)
  • Bianca (sensitive data)
    • Miarka (new for LifeScience)
• Storage
  • On-load directly connected to the clusters
  • Off-load for large data not needed for computation analysis anymore
• Cloud
  • Dis (region EAST-1)

High Performance Computing — HPC

What is a cluster?

• A network of computers, each computer working as a node.

• From small scale RaspberryPi cluster...

• To supercomputers like Rackham.

• Each node contains several processor cores and RAM and a local disk called scratch.

• The user logs in to login nodes via Internet through ssh or ThinLinc.

• Here the file management and lighter data analysis can be performed.

• The calculation nodes have to be used for intense computing.

Summary about the three "common" UPPMAX clusters

	Rackham	Snowy	Bianca
Purpose	General-purpose	General-purpose	Sensitive
# Nodes (Intel)	486+144	228+ 50 Nvidia T4 GPUs	288 + 10 nodes á 2 NVIDIA A100 GPUs
Cores per node	20/16	16	16/64
Memory per node	128 GB	128 GB	128 GB
Fat nodes	256 GB & 1 TB	256, 512 GB & 4 TB	256 & 512 GB
Local disk (scratch)	2/3 TB	4 TB	4 TB
Login nodes	Yes	No (reached from Rackham)	Yes (2 cores and 15 GB)
"Home" storage	Domus	Domus	Castor
"Project" Storage	Crex, Lutra	Crex, Lutra	Castor

Overview of the UPPMAX systems

  graph TB

  Node1 -- interactive --> SubGraph2Flow
  Node1 -- sbatch --> SubGraph2Flow
  subgraph "Snowy"
  SubGraph2Flow(calculation nodes) 
        end

        thinlinc -- usr-sensXXX + 2FA + VPN ----> SubGraph1Flow
        terminal -- usr --> Node1
        terminal -- usr-sensXXX + 2FA + VPN ----> SubGraph1Flow
        Node1 -- usr-sensXXX + 2FA + no VPN ----> SubGraph1Flow

        subgraph "Bianca"
        SubGraph1Flow(Bianca login) -- usr+passwd --> private(private cluster)
        private -- interactive --> calcB(calculation nodes)
        private -- sbatch --> calcB
        end

        subgraph "Rackham"
        Node1[Login] -- interactive --> Node2[calculation nodes]
        Node1 -- sbatch --> Node2
        end

Next session

We will try the different forms to log in to Bianca!

keypoints

• NAISS makes available large-scale high-performance computing resources, storage capacity, and advanced user support, for Swedish research.
• UPPMAX runs the local resources placed at Uppsala University
• A cluster consists of several inter-connected computers that can work individually or together.