Storage software GPU Instances

trainingmachineinfrastructure

Oracle Cloud Infrastructure (OCI) Compute provides industry-leading scalability and cost-performance for bare metal and virtual machine (VM) instances powered by NVIDIA GPUs for mainstream graphics, AI inference, AI training, digital twins, and HPC. OCI is the only major cloud provider to offer bare metal instances with NVIDIA and AMD GPUs for high performance that’s free of virtualization overhead. For checkpointing during AI training, our instances provide the most local storage per node (61.4 TB with H100 GPUs). For a balance of performance and price, OCI VMs with NVIDIA GPUs are consistently cheaper than AWS and Azure. High performance NVIDIA and AMD GPUs NVIDIA Tensor Core GPUs OCI offers the highest value and performance for bare metal and virtual machine compute instances powered by NVIDIA H100 Tensor Core GPUs, L40S GPUs, A100 Tensor Core GPUs, A10 Tensor Core GPU, and older-generation NVIDIA GPUs. OCI plans to offer instances with NVIDIA H200 and Blackwell GPUs. NVIDIA superchips OCI offers the NVIDIA GH200 Grace Hopper Superchip and plans to offer the GB200 Grace Blackwell Superchip for LLM inference. AMD Instinct GPUs OCI offers AMD Instinct MI300X GPUs with 192 GB of memory at a competitive price. High performance cluster networking Oracle’s ultralow-latency cluster networking, based on remote direct memory access (RDMA), provides microsecond-level latency. Superior GPU and infrastructure pricing Lower GPU pricing around the world Competing GPU instances from AWS and Azure can be consistently more expensive. Block storage price advantage AWS, Azure, and Google Cloud Platform can be up to 6X more expensive.