High performance computing

Powerful, flexible infrastructure to support scalable workloads.

Move at the speed of ideas

High performance computing is all about scale and speed. And when you're backed by Google Cloud’s state-of-the-art infrastructure, you can accelerate your most complex HPC workloads’ time to completion. Competitive pricing and discounts help you stay within budget to convert an idea into a discovery, a hypothesis into a cure, or an inspiration into a product.

Pay for what you use

Google Cloud offers on-demand access to custom machine configurations that are billed on a per-second basis with committed use discounts and sustained use discounts available. Plus, get up to 80% cost-savings with Preemptible VMs.

Innovate with high-performance, on-demand hardware

Access the latest Intel processors, NVIDIA GPUs, and high-throughput, low-latency storage with GCS. Build more across high-performance, on-demand hardware, CPUs, GPUs, and solid-state drives.

Cut the queues

On Google Cloud, each team has access to their scalable, tailor-made cluster. This helps them solve problems faster, reduce queue times for large-batch workloads, and relieve compute resource limitations.

HPC reference architectures

Preemptible VMs

Reference architecture for compute instances of batch jobs and fault-tolerant workloads.

Application scaling

Reference architecture for adding autoscaling, backup head node, monitoring, and alerting components.

Burst from on premises

Reference architecture for easily adapting and integrating on premises workloads to Google Cloud.

Industry use cases

From energy and manufacturing to government and climatology, global industries run their most demanding HPC workloads on Google Cloud.

Healthcare & Life Sciences

Powering medical breakthroughs.

See how Google Cloud and Big Data tools can process petabytes of genomic, imaging, clinical, and insurance data in seconds.

Financial Services

Bringing the numbers into view.

Power quantitative research, portfolio optimization, and risk modeling on Google Cloud — capable of running clusters with a few hundred (or thousand) cores in minutes.

Media & Entertainment

Render faster. Broadcast on demand.

Render large workloads, live stream videos to a global audience, and launch applications with the latest NVIDIA Tesla GPUs, even at peak times.

Our customers

The Broad Institute

This visionary research institution utilizes GCP to analyze human genomes 400% faster than on-site computing clusters — and at scale.

QuantConnect

Facing stability issues, an open-source financial trading community migrated to GCP to support the work of their 50,000+ quants.

Digital District

An award-winning French post-production company relies on GCP to handle massive computations without breaking the budget.

Resources

See Google Cloud high performance computing in action with these tutorials, solution examples, and videos.

Using clusters for large-scale technical computing in the cloud

High throughput computing with HTCondor

Deploy an Auto-Scaling HPC Cluster with Slurm

Using Cloud Pub/Sub for long-running tasks

File servers on Google Compute Engine

Next 2017 HPC/batch computing talk

Partner stories

Learn how our high performance computing capabilities are driving innovation for our partners.

Announcing integration of Altair HPC applications with Google Cloud

Easy HPC clusters on GCP with Slurm

NVIDIA® and Google Cloud Platform

Deploying the Elastifile cross-cloud data fabric

Google Cloud

Get started

Learn and build

New to GCP? Get started with qualifying GCP products for free with a $300 credit.

Need more help?

Our experts will help you build the right solution or find the right partner for your needs.