High Performance Computing (HPC)
High performance computing (HPC) on Oracle Cloud Infrastructure (OCI) brings powerful, cost-effective computing capabilities to solve complex mathematical and scientific problems across industries. OCI's bare metal servers coupled with Oracle’s cluster networking provide access to ultra-low latency RDMA (< 2 μs latency across clusters of tens of thousands of cores) over converged Ethernet (RoCE) v2.
HPC on OCI rivals the performance of on-premises solutions with the elasticity and consumption-based costs of the cloud, offering on-demand potential to scale tens of thousands of cores simultaneously. With HPC on OCI, you get access to high-frequency processors; fast and dense local storage; high throughput, ultra-low latency RDMA cluster networking; and the tools to automate and run jobs seamlessly.
Watch the video (2:42)
Performance of Ansys Mechanical on OCI
See how OCI compares with leading dedicated on-premises hardware systems for Ansys Mechanical simulations. Explore multiple options and choose the one that best fits in your cloud estate.
Deploy HPC workloads fast with OCI
Learn why high performance computing (HPC) on Oracle Cloud Infrastructure is the right solution for your HPC workloads and how easy it is to deploy HPC compute cluster through the Oracle Cloud marketplace.
Experience on-premises-level performance and control with OCI
Learn more about Oracle’s HPC solution that offers leading-edge compute, storage, networking, and allied software technologies at a lower cost while delivering the on-premises level of performance.
HPC customers and partners
Toyota moves high-performance workloads to Oracle Cloud
“We now run our HPC workloads on Oracle Cloud Infrastructure as part of our HPC multicloud strategy. OCI has incredible performance, and running computational fluid dynamics simulations with it has allowed us to improve the speed of computations and to optimize costs. This is helping us make the development of cars at Toyota more efficient, and produce cars with better performance.”
Shinichi Noda
Group Leader, DX Promotion Division, Toyota Motor
Red Bull Racing Honda and Oracle Partner to Elevate Data Analytics in Formula 1
“Races are decided by the narrowest of margins, but when it came to deciding who we wanted to work with, no other cloud company came even close.”
Christian Horner
Team Principal, Red Bull Racing Honda
Altair tests and runs complex simulations overnight on OCI
“We looked for the best price-performance, security, and VMware integrations. We found that in Oracle Cloud Infrastructure.”
Sam Mahalingam
Chief Technical Officer for Enterprise Solutions, Altair
Nissan gets to market quicker with Oracle Cloud HPC
“We selected Oracle Cloud Infrastructure’s HPC solutions as a part of our multi-cloud strategy to meet the challenges of increased simulation demand under constant cost savings pressure. I believe Oracle will bring significant ROI to Nissan.””
Bing Xu
General Manager, Engineering Systems Department, Nissan Motor Co, Ltd
Oracle Cloud Infrastructure Offers NVIDIA Quadro Virtual Workstation Image
“Around the globe, virtualization is helping enterprises stay productive during these challenging times. With Quadro Virtual Workstations on Oracle Cloud, creative and technical professionals can easily access the performance they need to work anywhere.”
Anne Hecht
Senior Director of virtualization product marketing, NVIDIA
BJSS Boosts High Performance Computing by 5x on Oracle Cloud
“Oracle Cloud Infrastructure is able to support deterministic latencies at the 10μs level at very high message volumes. There is sufficient evidence to justify exploring deployment of low-latency sensitive applications to OCI. This is significant because services requiring this service level avoid expensive on-site deployments.”
Larry Ryan
Chief Technical Officer, BJSS
YellowDog pioneers superfast HPC rendering jobs using Oracle Cloud
“Oracle’s bare metal compute and cluster networking technologies allowed BeeGFS on Oracle Cloud to outperform our on-premises HPC file system latency and throughput for MPAS workloads at a very low price point. Using Oracle Cloud’s RDMA cluster networking, BeeGFS can see performance of up to 140 GB/s with as little as 14 servers.”
Simon Ponsford
Chief Technical Officer, YellowDog
First principles: superclusters with RDMA—ultra-high performance at massive scale
Pradeep Vincent, Senior Vice President and OCI Chief Technical Architect
Jag Brar, Vice President and Distinguished Engineer
OCI architects break down how they applied RDMA networking that uses NVIDIA ConnectX RDMA network interface cards when building large-scale RDMA superclusters to support tens of thousands of GPUs.
Read the complete postFeatured blogs
- February 14, 2023 What’s Next: Training large language AI models at quintillions of operations on OCI
- December 13, 2022 First principles: Building a high-performance network in the public cloud
- AUGUST 30, 2022 Announcing dynamic performance scaling with OCI Block Volume auto-tuning
- JULY 6, 2022 Explore performance using SQL Explorer in Oracle Cloud Infrastructure Operations Insights
HPC reference architectures
Customers choose HPC on OCI
Leading price-performance
Oracle Cloud Infrastructure offers leading price-performance for HPC solutions. OCI's bare metal servers coupled with Oracle’s cluster networking provides access to ultra-low latency RDMA (< 2 μs latency across clusters of tens of thousands of cores) over converged Ethernet (RoCE) v2. With these capabilities, customers get optimized performance and scale for the most demanding HPC workloads at a lower cost. In addition, Oracle's flat, two-tier network topology provides uniform bandwidth and latency across all the nodes allowing HPC clusters to scale at flat prices.
Automated RDMA cluster deployment
Automate deployment of ultra-low latency RDMA clusters with Terraform template available on Oracle Cloud Marketplace stack to speed up running of performance-intensive workload across multiple machines. This template provides the base for HPC for weather modeling, CFD simulations, computational chemistry, genomics analysis, and many other use cases. The stack also provides the ability to install the Slurm scheduler, OpenMPI, and tools to test MPI connectivity with ease.
Easy AI/ML and data science
With Oracle, customers can effectively run AI/ML workloads on Oracle’s HPC infrastructure, including fast and large volumes of storage and computing power. To further support the data science teams, Oracle offers Oracle Cloud Infrastructure Virtual Machines for Data Science. This preconfigured environment enables customers to build models and deliver business value faster. OCI Virtual Machines for Data Science Service offers exceptional performance, security, and control. With OCI, customers can expand their compute resources as needed using compute autoscaling while keeping their costs under control by stopping compute instances when those are not required.
In addition, these virtual machines come with an NVIDIA GPU that can be up and running in less than 15 minutes with preinstalled common IDEs, notebooks, and frameworks. Oracle Cloud Infrastructure VMs for Data Science also includes basic sample data and code for data science teams to test and explore.
Run HPC workloads on a broad range of compute shapes
Oracle Cloud Infrastructure provides several compute options for bare metal and virtual machines instances from Intel, AMD, NVIDIA, and Ampere CPU/GPU based architectures.
Rich HPC partner ecosystem
Oracle works closely with an array of commercial and open source projects to support our customers. These partners test and build software on Oracle Cloud Infrastructure, which results in improved solution designs for customers across an array of computational tasks.
OCI Architecture Center for HPC
Your destination for guidance and technical resources for HPC on OCI.HPC technology center on OCI
High core frequencies and cluster networking give Oracle’s bare metal compute instances significant performance improvements over other public clouds and onsite data centers. Bare metal compute instances provide exceptional isolation, visibility, and control.
Our standard bare metal servers include dual 25 Gbps Ethernet for fast networking. In addition, Oracle’s back-end network fabric that uses Mellanox’s ConnectX-5, 100 Gbps network interface cards is groundbreaking. It provides RDMA over converged Ethernet (RoCE) v2, creating clusters with the same low-latency networking and application scalability as on-premises clusters.
Oracle Cloud offers high performance virtual machines, bare metal servers, and GPUs for graphics-intensive workloads for rendering, AI, and deep learning workloads.
Oracle Linux for High Performance Computing combines a fully supported, open, and complete operating environment that is 100% application binary-compatible with Red Hat Enterprise Linux. Oracle Linux delivers virtualization, management, and cloud native computing tools—along with the Linux operating system (OS)—in a single offering that meets high performance computing requirements.
Traditional storage simply can’t provide enough throughput for performance-intensive workloads. To meet these needs, Oracle makes it easy to deploy GlusterFS, BeeGFS, Lustre, and IBM Spectrum Scale high performance file systems that can deliver up to 453 GBps aggregate throughput to HPC clusters.
Oracle Marketplace gives our customers the ability to evaluate, select, and deploy end-to-end solutions seamlessly while automating the deployment process for HPC solutions.
Get hands-on with HPC on OCI
HPC solutions by industry
Oracle built the infrastructure and services in the cloud to support enterprise-class customers' needs across multiple industries and industry use cases.
Manufacturing
HPC on OCI allows our manufacturing customers to run larger workloads faster and support more computationally complex simulations. From crash-test simulations to aerodynamics, HPC on OCI provides the compute capability to run these workloads at a higher performance than any other cloud provider.
Computational Fluid Dynamics (CFD)
CFD is a common workload that simulates the motion of air and fluid to simplify and speed product engineering. For example, in the automotive sector, it helps manufacturers simulate cabin airflow, engine oil dynamics, and the air flow around the car to improve fuel efficiency. It is a tightly coupled MPI-based workload that benefits from Oracle’s 100 Gbps cluster networking, high-frequency Intel processor-based compute instances, and the latest NVIDIA GPUs.
Nissan Moves to Oracle Cloud Infrastructure for High-Performance Computing
CFD is a common workload that simulates the motion of air and fluid to simplify and speed product engineering. For example, in the automotive sector, it helps manufacturers simulate cabin airflow, engine oil dynamics, and the air flow around the car to improve fuel efficiency. It is a tightly coupled MPI-based workload that benefits from Oracle’s 100 Gbps cluster networking, high-frequency Intel processor-based compute instances, and the latest NVIDIA GPUs.
Life sciences simulations on OCI
Molecular dynamics (MD) and genomic simulations are routine workloads of the life sciences industry. These simulations analyze the physical movements of atoms and molecules and are utilized in use cases like drug discovery. By moving these computationally intensive workloads to OCI, researchers can achieve the best possible performance, meet scale requirements, and reduce the time for discovering new treatments, which can offer cost savings.
Financial Services
Financial applications, including trading applications, require high-performance, low-latency infrastructure. These applications were not a design goal of early cloud architectures, and they have been slow to move to the cloud. Oracle Cloud Infrastructure provides the performance characteristics, such as sub-2-microsecond in-cluster latency these applications require, rivaling custom-built and expensive on-premises solutions.
Visual effects rendering
High performance computing provides the horsepower for today’s omnipresent visual effects. From special effects in movies to TV ads and the latest PC and console game titles, all are developed by media companies who need HPC and GPU performance on-demand. OCI offers HPC and GPU-powered bare metal instances like NVIDIA Quadro Virtual Workstation that deliver consistent performance in line with expensive high-end graphics workstations at a lower cost.
Higher Education & Research
Oracle Cloud Infrastructure’s supercomputing platform gives researchers access to bare metal NVIDIA GPUs, high performance computing instances, and a low-latency clustered network. Researchers can create clusters for running large-scale computations to accelerate the research in multiple branches of science and engineering like drug discovery, genomics, weather forecasting, space exploration, and more. Through programs like Oracle for Research, Oracle is working closely with research organizations like the University of Bristol and the Royal Holloway University of London to help accelerate the development of vaccines and advanced solutions that address climate change.
The Oracle Playbook series
We've compiled the secrets to our people, process, and systems strategy. And we want to share it with you.
Resources
Press releases
Blogs
- Running NVIDIA Clara Parabricks Pipelines on Oracle Cloud Infrastructure
- Run high-performance computing workloads anywhere in the world with Oracle Cloud
- Efficiently managing cluster network resources on Oracle Cloud using the cluster network resize feature
- Bring Ansys Fluent to new heights with Intel on Oracle Cloud Infrastructure
- Advancing healthcare with high-performance computing on Oracle Cloud Infrastructure
Industry articles
Videos
- Oracle Live: Discussion with Jensen Huang, Founder and CEO, NVIDIA (4:24)
- Oracle Live: Update from Lisa Su, CEO and President, AMD (1:22)
- Oracle Live: Update from Bob Swan, CEO, Intel (2:43)
- Oracle Live: Discussion with Renee James, Chairman and CEO, Ampere (3:46)
- Nissan Gets to Market Quicker with Oracle Cloud HPC (1:18)