Stay organized with collections Save and categorize content based on your preferences.
Register for Next '22 at no cost and add HPC sessions to your playlist. Register now!
Jump to

High performance computing

Google Cloud’s high performance computing (HPC) solutions are easy to use, built on the latest technology, and cost-optimized to provide a flexible and powerful HPC foundation that clears the way for innovation. 


A flexible and powerful infrastructure that clears the way for innovation

Get results faster

Create HPC clusters in minutes based upon proven best practices with HPC Toolkit. Quickly start jobs with predictable performance with pre-configured HPC VMs

Enable and sustain innovation

Enable your team with high performance on-demand resources. Build and run your most intensive workloads on the latest Intel and AMD CPUs, NVIDIA GPUs, and high-throughput, low-latency object and file storage.

Pay only for what you need

Control and manage costs as you scale with committed and sustained use discounts. Get up to 90% cost savings by using Spot VMs for your most flexible workloads.

Related services


See tutorials, guides, and resources for this solution

Best Practice
Best practices for running tightly coupled HPC applications

Learn best practices for tuning Google Cloud resources for optimal Message Passing Interface (MPI) performance.

Easily deploy HPC environments with Google Cloud's Toolkit

For example, use client libraries to create and manage Compute Engine resources in Go, Python, Java, Node.js, and other languages.

Parallel file systems for HPC workloads

Review the storage options in Google Cloud for HPC workloads and learn when to use parallel file systems like Lustre and DDN EXAScaler Cloud.

Lustre file system in Google Cloud using DDN EXAScaler

Learn how to design a Lustre file system for HPC workloads. Get an overview of the process to deploy a Lustre file system in Google Cloud by using DDN EXAScaler.

Cloud HPC Toolkit

Cloud HPC Toolkit documentation.

Genomics: Running a GATK best practices pipeline

Run a pipeline on Google Cloud using the GATK Best Practices for variant discovery in whole genome sequencing data provided by the Broad Institute.

Google Cloud Public Datasets

Explore high-demand public datasets, making it easy for you to access and uncover new insights in the cloud.

Create a virtual GPU-accelerated Linux workstation

Create a virtual Linux workstation running CentOS with one of three attached display-capable NVIDIA GPUs T4, P4, and P100.

Deploy a Slurm cluster on Compute Engine

This tutorial shows how to deploy a Slurm cluster on Compute Engine. The Slurm Resource Manager is a popular resource manager used in many HPC centers.