GPU instances
Oracle Cloud Infrastructure (OCI) Compute provides bare metal and virtual machine instances powered by NVIDIA GPUs for a variety of use cases, including mainstream graphics and videos as well as the most demanding AI training and HPC workloads. OCI excels at both training times and cost because of its ultralow latency and near line-rate network performance.
OCI at NVIDIA GTC, the conference for AI and the metaverse
GPU instances—key features
Remote direct memory access (RDMA) communication between instances supports large GPU clusters with 1,600 Gb/sec of bandwidth and ultralow latency.
High performance NVIDIA GPU instances
NVIDIA Tensor Core GPUs
OCI offers the highest value and performance for bare metal and virtual machine compute instances powered by these NVIDIA GPUs: A100, A10, V100, and P100.
Bare metal performance
Bare metal compute instances allow customers to run high performance, latency-sensitive, and specialized workloads directly on server hardware—just as they would on-premises.
High performance cluster networking
Oracle’s ultralow latency cluster networking, based on RDMA, provides microsecond-level latency. Without compromising on performance, customers can host more than 500 GPUs in a cluster.
Choose from a variety of VM and bare metal compute instances
Virtual machines and bare metal
Match your workload against different instance types to find your ideal fit.
VM instances
For VMs, choose from NVIDIA Ampere, Volta, and Pascal GPUs with 1 to 4 cores, 16 to 64 GB GPU memory, and up to 480 GB of locally attached flash storage.
Bare metal instances
Cluster bare metal instances for HPC and AI training using NVIDIA A100 Tensor Core GPUs with 640 GB of GPU memory per node.
Reduce networking and storage costs
Consistent global pricing
OCI’s pricing is simple and transparent across regions, ensuring that cloud spending is predictable regardless of the location where your workloads run.
Secure cloud infrastructure
Isolated network virtualization
Prevent attacks on customer tenancies with isolated network virtualization. A foundational element of Oracle Cloud Infrastructure’s security-first architecture, the design stops malware in its tracks with a custom-designed SmartNIC to isolate and virtualize the network.
Oracle Cloud Infrastructure Identity and Access Management
Using easy-to-define policies organized by logical groups of users and resources, control not only who has access to Oracle Cloud Infrastructure resources but which ones and the access type. Customers manage identities and grant access by leveraging existing organizational hierarchies and federated directory services, including Microsoft, Okta, and other SAML directory providers.
Hardware root of trust
Reduce the risk of firmware-based attacks against Oracle Cloud Infrastructure customer tenants with custom-built, hardware-based, root-of-trust-technology designed to wipe and reinstall the firmware every time a new server is provisioned or a new customer tenancy is established.
Oracle Web Application Firewall
Protect applications from malicious and unwanted internet traffic with a cloud-based, PCI-compliant, global web application firewall (WAF) service. By combining threat intelligence with consistent rule enforcement, this service strengthens defenses and protects internet-facing application servers.
Oracle Cloud Infrastructure Vault
Centrally manage and maintain control of the encryption keys and secret credentials used to protect enterprise data and access resources. The security for both are stored in Federal Information Processing Standards (FIPS) 140-2, Level 3-certified, hardware security module (HSM).
Users can further control permissions for individual keys and vaults with OCI Identity and Access Management and monitor key lifecycle changes with OCI Audit to meet enhanced compliance requirements.
Choice of operating systems
Choose from popular operating systems
Oracle supports an extensive list of Microsoft Windows Server and enterprise Linux operating systems, including Oracle Linux, CentOS, Ubuntu, and many others. Oracle Autonomous Linux is a free, autonomous operating environment that reduces complexity and human error to deliver increased cost savings, security, and availability.
Automate operating system management
The OCI OS Management service allows customers to set policies for automated updates and patches for their operating system environment.
Bring your own image (BYOI)
Customers can run their own operating systems and hypervisors on Oracle compute instances and use the same image across configuration sizes.
Access readily-available software
Access software and disk images
Oracle Cloud Marketplace provides software and disk images for data science, analytics, artificial intelligence (AI), and machine learning (ML) models so customers can quickly gain insight from their data.
NVIDIA GPU Cloud
Access hundreds of GPU-optimized applications for machine learning, deep learning, and high performance computing covering a wide range of industries and workloads. Learn more.
NVIDIA RTX Virtual Workstation
Customers can deliver powerful workstation performance wherever employees need it by running NVIDIA RTX Virtual Workstation on Oracle Cloud.
Microservices and containers
Container registry
Developers building applications using containers leverage a highly available, Oracle-managed private container registry service for storing and sharing container images. Push or pull Docker images to and from the registry using the Docker V2 API and the standard Docker command line interface (CLI). Images can be pulled directly into a Kubernetes deployment.
Oracle Functions
Functions as a service (FaaS) lets developers run serverless applications that integrate with Oracle Cloud Infrastructure, Oracle Cloud Applications, and third-party services. Gain developer efficiency along with the community of the open source Fn Project.
Talk with Oracle about accelerating your GPU workloads.
GPU instances—use cases
AI infrastructure for deep learning training and inferencing
Train AI models using OCI Data Science, bare metal instances, cluster networking based on RDMA, and NVIDIA GPUs.
This diagram describes two stages of deep learning model development: model training and model inferencing. In model training on the left, the untrained neural network is input to a training algorithm enabled by OCI Data Science, bare metal compute, local storage, and cluster networking. The output of the training algorithm is a trained model with a new capability. The model inferencing step is described on the right. Consider a trained model such as DALL-E 2, which can take text inputs and generate images. A text input is fed into the trained model, and an image output from the model is provided.
Virtual desktop infrastructure (VDI)
OCI Compute powered by NVIDIA GPUs provide consistent high performance for VDI.
Virtual desktop infrastructure
CFD and high performance computing using GPU instances
OCI enables computer-aided engineering and computational fluid dynamics for fast predictions of the aerodynamic properties of objects.
CFD and high performance computing using GPU instances
