Run high-performance computing workloads anywhere in the world with Oracle Cloud

July 28, 2021 | 4 minute read
Jigar Modi
Principal Technical Program Manager
Text Size 100%:

Having access to the latest and greatest technology with a wide selection of hardware to match each workload is central to the value proposition of Oracle Cloud Infrastructure (OCI). OCI also provides enterprise-grade production high-performance computing (HPC) capability and the most cost-effective pricing in the public cloud. Our core tenet has always been the absolute best price-performance, giving customers the benefits of on-premises deployments with all the flexibility of public cloud.

OCI performance derives from Oracle’s rigorous approach to HPC, running the truest bare metal offering with the most-performant, lowest-latency HPC network of any major cloud. This combination allows tightly coupled HPC applications, whose performance hinges on the application’s ability to communicate between nodes with minimal delay, to scale to larger core counts without losing efficiency, saving time and money.

OCI is the true premiere cloud for all your HPC needs. With more locations of specialty hardware than any other cloud provider, you can run your workloads anywhere and at any time. OCI has true disaster protection with compliance in all regions with dual cloud regions in six countries and EU today. Our regions have up to four availability domains and three fault domains per availability domain.

Why should you switch to Compute optimized X9 instances today?

We’re launching the X9 Optimized instances, which on based on Intel Ice Lake offerings and have the highest performance per core.

These enhanced instances cater to the need for the customers who value the extra performance for compute-hungry workloads, such as computational fluid dynamics (CFD), crash simulations, and 3D visualization. The magic behind performance is Intel’s Ice Lake generation, their 10-nm offering with better PCIe bandwidth, greater per-socket performance (25% ISO frequency), and greater error correction capabilities compared to Skylake. Electronic-design analysis and Monte Carlo simulations are expected to gain more than 40% performance, compared to the existing X7 generation of instances. The other key benefit of X9 is the favorable pricing, which is offered on par with X7 shapes.

These Intel-powered Compute instances are the first to feature our flexible Compute capabilities, which enable you to choose any granular number of cores and amount of memory to suit your workload’s needs. You’re no longer limited to fixed virtual machine (VM) sizes with four, eight, or 16 cores and static memory ratios per core. You can deploy a five-core VM with 25 GB of RAM and pay only for what you use.

Currently, we’re offering X9 instances in US, EMEA, and JAPAC regions and continue investing to ramp up more hardware to expand our global footprint in HPC platform.

Our customers have seen performance improvement of around 18% to 30% on X9 even when using older versions of applications and MPI.

Latest hardware for demanding workloads

For of GPU, Oracle is introducing methods and instances working with NVIDIA’s A100 accelerator. We gauged the need and are adding more racks across regions in coming weeks to meet the aggressive demand for our A100 NVIDIA GPUs. We have A100 available in 17 out of 29 regions all around the world at an industry-low, on-demand price of $3.05 per GPU per hour. This bare metal instance joins our cluster network architecture, allowing customers to scale up to 512 GPUs in a single cluster for large-scale artificial intelligence training or HPC workloads.

These instances provide up to 1.6 TB per second of bandwidth per bare metal node that houses eight A100 GPUs, all fully interconnected with NVLINK. These instances also provide you with over 25 TB of local NVMe storage and 2 TB of RAM for large-scale graph workloads or accelerated databases. Finally, we also allow for cutting edge capabilities such as GPU Direct through RDMA, not yet available on any other cloud provider.

Spearheading this newest period for OCI with updating their offerings to consist of the most current x86 processors from AMD and Intel, Oracle also plans to start their initial Arm-driven scenarios dependent on Ampere’s Altra processors.

In short, Oracle is rolling out the latest and finest in Compute hardware to woo other cloud consumers, maintain up with the competitiveness, and usually entice traditional, on-premises shoppers that it’s time to come to the cloud.

To learn more, see NVIDIA and Oracle Cloud Infrastructure NVIDIA GPU Cloud Platform.

A solution that fits your workloads

Start your 30-day free trial and get access to a wide range of Oracle Cloud services for 30 days. If you want more guidance, get hands-on experience with our cloud on our LiveLabs platform.

Learn more about running HPC workloads on Oracle Cloud Infrastructure with the following documentation:

Jigar Modi

Principal Technical Program Manager

Previous Post

Multicloud solution using Oracle Exadata Cloud service for data management with applications running on Azure

Niranjan Mohapatra | 9 min read

Next Post

Oracle Cloud Infrastructure Data Integration’s latest release improved productivity

Sandrine Riley | 5 min read