WHY THIS MATTERS IN BRIEF
Why buy one giant supercomputer when you can just pool together the power of many smaller systems to beat the biggest supercomputers?
Love the Exponential Future? Join our XPotential Community, future proof yourself with courses from XPotential University, connect, watch a keynote, read our codexes, or browse my blog.
It used to be the case that organisations would either spend huge sums of money buying their own supercomputers and exascale supercomputing platforms, or they wouldn’t and they wouldn’t have access to the raw computer power these monsters have. Then supercomputing moved into the cloud, and more people could access them. But, now there are more companies, like Microsoft with their Azure cloud platform, who are finding new ways to aggregate together small clusters of supercomputers and the infrastructure of giant hyperscale cloud datacenters to create a new class of so called “federated” supercomputers that make many of today’s most powerful supercomputers look like toy calculators.
As one of the lead companies in this space IT services business Atos this week introduced a suite of cloud-based high-performance computing (HPC) services, based around technology gained from its purchase of cloud provider Nimbix last year.
The Nimbix Supercomputing Suite is described by Atos as a set of flexible and secure HPC solutions available as a service. It includes access to HPC, AI, and quantum computing resources, according to the services company.
In addition to the existing Nimbix HPC products, the updated portfolio includes a new federated supercomputing-as-a-service platform and a dedicated bare-metal service based on Atos BullSequana supercomputer hardware.
The latter, the Nimbix Dedicated Supercomputing service, provides customers with dedicated bare-metal access to Atos BullSequana X servers, or bare-metal-as-a-service (BMaaS) as the company brands it.
However, Atos does not make it clear exactly which of its line-up of systems will be part of this service. The recently launched XH3000 systems support AMD Epyc processors in addition to Intel Xeon and Nvidia GPUs, as detailed by our sister site The Next Platform.
Meanwhile, the Nimbix Federated service is pitched by Atos as a federated supercomputing-as-a-service, which it claims will provide customers with a unified console to manage HPC resources across both private ecosystems and public clouds. Both public and private infrastructure operators will be able to take advantage of each other’s resources, Atos said, with the unified service console managing all compute zones and regions involved.
Emmanuel Le Roux, Atos head of HPC AI & Quantum, said the new suite would provide the performance and cost-effective advantages that enterprises are seeking for their data-driven environments.
“Also, thanks to our longstanding expertise in multi and hybrid cloud built into our newly launched offerings, customers can integrate their HPC workflows in the most comprehensive as-a-service portfolio in the industry, including seamless integration with any hyperscale CSPs,” Le Roux added.
Addison Snell, CEO of HPC analyst outfit Intersect360, said in a statement: “The range of Nimbix Supercomputing Suite allows organizations to choose the type of cloud-immersed deployment that makes the most sense, depending on the workflow, whether it’s utility computing on-demand, federating workloads across multiple clouds, or simply converting on-premises systems to cloud models of accounting and operations.”
The introduction of the Nimbix Supercomputing Suite is the sign of a move away from focusing on traditional infrastructure services for Atos in favor of the cloud after the company’s revenues shrank by 2.6 percent last year.
