In this hands-on tutorial, I'll walk you through the process of deploying a Slurm workload manager on Google Cloud. Perfect for machine learning and high-performance computing workloads!
Free Trial - Our New Diagram Tool: https://softwaresim.com/pricing/ ("YOUTUBE24" for 25% Off)
Demonstration Diagram: https://github.com/nodematiclabs/slurm-setup
What You'll Learn:
- Setting up a new Google Cloud project for Slurm
- Enabling necessary Google Cloud APIs (Filestore, Cloud Storage, Service Networking)
- Using the Google Cloud Cluster Toolkit to deploy Slurm
- Understanding the blueprint architecture with debug nodes, compute nodes, and H3 nodes
- Connecting to and running jobs on your Slurm cluster
- Proper teardown to avoid unnecessary costs
Have questions about setting up Slurm on Google Cloud? Drop them in the comments below!
0:00 Conceptual Overview
1:00 Service APIs
2:35 Service Accounts
3:32 Cluster Toolkit
4:59 Cluster Blueprint
8:14 Create and Deploy
13:01 Slurm Login Node
#googlecloud #HPC #machinelearning
Download
0 formats
No download links available.
Slurm Setup Made Simple (Cluster Toolkit and Google Cloud AI/ML/HPC) | NatokHD