Over the last year we’ve continued to make Oracle Cloud one of the the best platforms for high-performance computing (HPC) workloads. We designed our cloud with HPC as one of the core use-cases, and this influenced everything from our choice of server hardware to our data-center design with nonblocking network to ensure low-latency and high-bandwidth connectivity between compute nodes. Our new managed File Storage service is built with performance as one of its most important characteristics, enabling you to offload the management of a high-performance clustered file system.
We also introduced instances powered by NVIDIA’s Tesla V100 GPUs released at NVIDIA’s GPU Tech Conference in March. Initially these instances were generally available only in our Ashburn, VA, data center in bare metal instance types. Today, we’re excited to announce the expansion of these instances to our London region along with the preview availability of virtual machine shapes allowing customers to get instances based on Intel® Xeon® Scalable Processors with 1, 2 or 4 NVIDIA Tesla V100 Tensor Core GPUs. You can find more information on our NVIDIA GPU microsite.
We understand how important the ISV and Partner ecosystem is to the HPC community. We’ve been making steady progress through our previously announced partnerships with important global ISVs such as Altair, Citrix, and Teradici.
Today we’re excited to announce a partnership with one of the global leaders in engineering simulation, ANSYS. ANSYS offers a portfolio of engineering simulation products to help customers solve the most complex design challenges and engineer products limited only by imagination.
“The partnership with Oracle Cloud Infrastructure enables our customers to run simulations in a bare metal environment with the same world-class experience and consistency that they have on-premises,” said Wim Slagter, Director HPC & Cloud Alliances at ANSYS. “The partnership expands on ANSYS' Open Cloud Strategy—empowering customers with flexibility to run simulations on their cloud platform of preference.”
Read more about our work with ANSYS, including initial benchmarks and performance testing on Oracle Cloud Infrastructure. Over the course of the year, you’ll see benchmarks, easy-to-deploy templates, and other fantastic tools to make it easy to deploy and run HPC software on Oracle Cloud Infrastructure.
Through our 20-year collaborative and impactful partnership with Intel, we’re continuing to accelerate our HPC efforts. Our recent work on X7 Intel Xeon Scalable Platinum instances was released at the last Oracle OpenWorld, and longer-term work is occurring on projects such as Oracle in-memory databases and Oracle Exadata.
“Intel’s innovative HPC foundation, based on the Intel® Xeon® Scalable processors, includes critical platform innovations in memory, storage and acceleration technologies to address the complex spectrum of diverse HPC workload requirements. We’ll continue to work with Intel over the course of the year with some exciting announcements relating to HPC later in the year”, said Vinay Kumar, Vice President of Product Management, Oracle Cloud Infrastructure
HPC is a major area of investment for us. We’re trying to solve the challenges that customers face with both cloud- and on-premises based deployments in which compute and network performance is a major challenge for tightly coupled or MPI workloads.
“Intel has a long history working with Oracle on offerings that are optimized for enterprise workloads. We’re excited about our collaboration with Oracle Cloud Infrastructure, which started with the initial launch of X5 Compute Instances in 2016, and then the launch of X7 Compute Instances at the end of 2017. Intel continues to work closely with Oracle Cloud Infrastructure on further collaborations on HPC where users will experience significant performance improvements in workloads compared to previous generation hardware” said Jeff Wittich, Director, Business Strategy Platform Enablement, Intel
Enterprises today are using our HPC capabilities in our various regions around the globe. Customers such as YellowDog and Zenotech are already benefiting from these super-computing capabilities. YellowDog provides a platform that enables animation studios and VFX facilities to access tens of thousands of cores or GPUs to deliver intensive rendering workloads within seemingly impossible deadlines. Zenotech provides a simulation-as-a-service platform that enables customers to run computational fluid dynamics. Customers find us to be a consistently faster, cheaper, and more efficient option compared to other cloud providers or even on-premises clusters.
We’ll be demonstrating and talking about some of these HPC use cases in our sessions next week, as well as at our session Run Cloud HPC and GPU Applications Without a Virtual Layer on Wednesday, June 27, from 3 p.m. to 3:20 p.m. at booth N-210, and at a quick vendor showdown session.
You can also visit our booth at G-822 to meet our engineering teams, learn more, and get free credits to run your own HPC jobs and workloads on Oracle.
Looking forward to seeing you all there!