Private AI infrastructure is gaining traction as organizations seek greater control over sensitive data and operating costs while meeting regulatory requirements. ThinkOn Inc. is extending its VMware-based cloud model to meet that demand while keeping Canadian workloads under domestic operational and legal control.
The company’s sovereign cloud provides the infrastructure foundation for Canada’s Sovereign AI Platform, which serves thousands of federal employees across 23 departments, according to ThinkOn. The company is also seeing broader interest in private AI from organizations weighing data control and the cost of on-demand services, according to Craig McLellan (pictured), founder of ThinkOn.
“We’re a modest-sized company. We get between three and four inquiries a week,” he said. “I would say half of those, for AI-type services, are for private AI. We’re starting to see quite a significant bump in the level of activity.”
McLellan spoke with theCUBE Research’s Christophe Bertrand and co-host Alison Kosik at VMware Explore, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed sovereignty, compliance and the economics of operating private AI on VMware infrastructure. (* Disclosure below.)
Private AI infrastructure adds sovereignty and cost control
ThinkOn uses VMware Cloud Foundation as a common control plane for its established private cloud and newer AI services. The company combines that technology with Canadian data centers, domestically based engineers and documented compliance processes to address supply-chain and jurisdictional requirements, according to McLellan.
“Fundamentally, the test for sovereignty is, ‘Do I have unfettered access to the infrastructure on which I choose to operate, yes or no?’” he said. “If the answer is no, then you just cannot claim sovereignty.”
Customers are also evaluating private AI infrastructure as an alternative to variable-cost, on-demand services, according to McLellan. ThinkOn’s approach reflects a broader move toward running private AI on a unified VMware platform while matching graphics processing unit capacity to individual workload requirements.
“It’s not always about the latest and greatest GPUs,” McLellan said. “Not everyone needs B-series or H-series. We have people that are running quite comfortably on A-series GPUs, and they’re elated because they’re being rewarded financially for that. But the beautiful part about our private AI strategy is it’s the same control plane.”
Here’s the complete video interview, part of theCUBE’s coverage of VMware Explore:
(* Disclosure: TheCUBE is a paid media partner for the VMware Explore event. Sponsors of theCUBE’s event coverage do not have editorial control over content on theCUBE or SiliconANGLE.)
Photo: SiliconANGLE
