Why HPC shops have their heads in the cloud – HPCwire

As they face a growing demand for high performance computing to power modern applications, organizations are increasingly turning to cloud service providers.

Over the past few years, the concept of cloud-based high performance computing resources has come of age. In a recent report, Hyperion Research predicts that spending on HPC in the cloud will post strong growth over the next few years, to reach $9 billion in 2024.

“Cloud service providers (CSPs) are key to this growth as they are taking a stronger interest in the HPC market, focusing on delivering capabilities that more closely match on-premise systems in performance, security, and in some cases, price,” Hyperion notes.

Hyperion cites multiple drivers for this move to HPC in the cloud. For starters, cloud offerings have become much more capable for HPC applications than they were in the past. And at the same time, CSPs and other companies are providing easier access models to clouds, along with a larger set of applications available.

Another driving factor is that cloud platforms can be an important enabler in the rapidly emerging markets of artificial intelligence and other high performance data analysis applications. In these cases, HPC shops are looking to the cloud for resources to power demanding workloads for automated driving systems, precision medicine, affinity marketing, business intelligence, cyber security, smart cities, the internet of things and more.

“As HPC applications grow in complexity, as in the case with the addition of new AI techniques to a wide range of application portfolios, the ability to easily burst into the cloud provides a more flexible, responsive computing environment,” the firm notes. “Likewise, multicloud capabilities are increasingly appealing to many HPC sites as they allow users to manage multiple third-party cloud resources from a single ‘pane of glass,’ a feature especially attractive to global corporations that use different CSPs scattered across different geographic locations.

Meanwhile, many CSPs are moving to provide support specifically targeted for HPC workloads, adding in-house HPC experts, and offering services to help users better understand how to run their applications in the cloud, Hyperion says. With the added HPC expertise from CSPs, HPC users have more resources at their disposal to place the right instances on the right platform for each of their jobs.

A case study: Spire

Spire, one of the largest and most trusted space-to-cloud analytics companies uses satellite data and algorithms to provide what is arguably the most advanced maritime, aviation and weather tracking in the world.

To carry out its ambitious mission, Spire requires leading-edge HPC systems that can process large amounts of data at lightning-fast speed. For this need, Spire leverages HPC on demand from R Systems, a Dell Technologies partner. These cloud-based services give Spire’s engineers and data scientists untethered access to best-in-breed technologies, including bare-metal clusters, InfiniBand networking, parallel file systems, and other computing technologies.

HPC on demand remotely runs Spire’s advanced numerical weather predictions initialized with the radio occultation data collected by the company’s satellites. The combination of this unique data and powerful cloud-based HPC systems allows Spire to deliver high-quality weather prediction services to its customers in weather-sensitive markets. Read the case study.

Getting started

Organizations that are interested in understanding their opportunities to leverage HPC in the cloud don’t need to go it alone in their investigations. There are many resources available to help with the process.

One of these is an online Cloud Application Assessment Tool from Hyperion Research. This tool is designed to help engineers, IT professionals, HPC users and buyers, and business executives understand the extent to which their workloads are amenable to operation on a public cloud. The purpose of this tool is not to promote cloud-based options, but rather to provide an unbiased assessment of where an application could be best operationally suited.

Explore the diverse on-demand offerings in the Dell Technologies APEX portfolio. APEX delivers IT infrastructure solutions at subscription rates for a range of data and workload requirements, enabling IT organizations to gain easy access to the resources they need while staying in control of IT operations. Specific APEX options include storage services, cloud services and custom solutions.

These offerings are part of a broader Dell Technologies portfolio for AI, HPDA and HPC Hybrid Cloud, including subscription-based offerings.

Ultimately, HPC in the cloud has a tremendous amount of momentum, and there are many ways to get there. The key is to understand your needs and your opportunities for cost-effectively leveraging cloud-based resources.

For a closer look at the topics explored here, read the paper and see the on-demand webinar “Bringing HPC Expertise to Cloud Computing,” presented by Dell Technologies and Hyperion Research.

Spread the love

Leave a Reply

Your email address will not be published.