As enterprises across Asia prepare for the hybrid AI wave, Penguin Solutions sharpens its cloud-to-edge strategy with resilient systems and intelligent services.
Penguin Solutions, a rebranded entity emerging from the 2022 acquisition of Stratus Technologies by Smart Global Holdings (SGH), is positioning itself as a formidable player in the next generation of high-performance, fault-tolerant computing. With AI deployments spreading from data centres to factory floors, Penguin’s unified portfolio — spanning AI and HPC infrastructure, fault-tolerant edge systems, intelligent memory modules, and software-defined management — is targeting the hybrid computing demands of the future.
“By 2028, 75% of enterprise workloads will run in hybrid environments — from the core to the edge,” said Stephen Greene, Vice President of Global Marketing for Penguin’s Advanced Computing business, in a recent conversation with AsiaBizToday. “We’re moving well beyond centralised training of AI models to inferencing at the point of customer interaction or product creation—right at the edge.”
From Resilient Roots to AI-Driven Reach
Penguin’s pedigree runs deep. Stratus, with over four decades of expertise in fault-tolerant computing, remains core to Penguin’s identity. Its flagship platforms—ztC Endurance® and ztC Edge® — are known for “seven nines” (99.99999%) availability and are used extensively in sectors such as pharmaceuticals, transportation, oil and gas, and mining.
Meanwhile, the legacy Penguin Computing business brings cutting-edge AI and HPC infrastructure credentials, including partnerships with hyperscalers like Meta and voltage-intensive clients such as Voltage Park. As of August 2024, Penguin boasts $1.2 billion in FY24 revenue, with over 2,700 employees globally and nearly 90,000 GPUs deployed and managed across clients.
Greene noted that Penguin’s services go far beyond hardware. “We’re not in the business of shipping boxes. We design, build, deploy, and manage full infrastructure — from racks and GPUs to software and spares,” he said.
Asia Pacific Strategy: Focused Expansion, Pragmatic Rollout
While advanced AI and HPC infrastructure is gaining traction in North America and Japan, Penguin is adopting a more targeted approach in the Asia-Pacific region. “We’re not going all-out yet in APJ,” said Lin Hoe Foong, Vice President and Managing Director, APeJ. “We’re being selective — focusing on opportunities as they emerge, especially in places like South Korea, where we have a key partnership with SK Telecom.”
In Southeast Asia, the company sees strong demand for its legacy Stratus fault-tolerant solutions in sectors like manufacturing, transportation, and critical infrastructure. “We’ve been steadily expanding in the industrial automation space since 2018,” Lin Hoe said, noting applications in train networks, pharmaceutical plants, and offshore rigs.
Lin Hoe also pointed to growing opportunities in cybersecurity, especially as enterprises virtualise their security stacks. “As network function virtualisation picks up, the reliability of server infrastructure becomes critical,” he explained. Penguin is working closely with vendors like Fortinet to ensure uptime for virtual firewalls and identity management systems.
Repatriation from Cloud and Search for VMware Alternatives
With public cloud costs rising in several Southeast Asian markets, Penguin sees a trend of workload repatriation—shifting compute back to on-premise or private cloud environments. “As enterprises bring workloads back in-house, they need infrastructure that is both resilient and scalable,” Lin Hoe said.
Penguin also aims to capitalise on the growing dissatisfaction with VMware following its acquisition by Broadcom. “Customers want alternatives. They’re looking at Red Hat, Microsoft Hyper-V, and other hypervisors. We’re vendor-agnostic and well-positioned to support those transitions,” he added.
Edge AI: The Next Frontier
With AI inferencing at the edge poised to become a major trend, Penguin is gearing up to launch GPU-powered versions of its ztC Endurance boxes by late 2025. “These will support smaller-scale inferencing in remote data centres and even field environments,” Greene said, adding that the architecture will prioritise real-time decision-making with minimal latency.
Through its ICE ClusterWare™ and ICE ClusterWare AIM™ platforms, Penguin offers intelligent infrastructure software for managing large AI clusters and optimising GPU usage — capabilities already proven in engagements with Meta and Georgia Tech.
What sets Penguin apart is its service-led philosophy. With more than 20 years of experience in HPC and AI infrastructure, Penguin claims to help enterprises avoid the pitfalls of DIY experimentation. “Many companies are learning on the job, but CIOs shouldn’t have to pay for vendor mistakes,” Greene said. “Our role is to de-risk the infrastructure journey.”
As the lines between IT and OT blur, and AI spreads from cloud to the physical edge, Penguin Solutions is betting that its combination of rugged hardware, intelligent software, and long-term support will resonate with enterprises in transition.
“Whether it’s a bank in Singapore, a telecom operator in Korea, or a mining company in Australia — we’re helping customers simplify the complexity of modern computing,” Greene said.