A new wave of GPU-focused cloud providers is offering high-end hardware at prices lower than those charged by hyperscalers. Dedicated infrastructure needs to be highly utilized to outperform these neoclouds on cost.
Hyperscalers design their own servers and silicon to scale colossal server estates effectively. AWS uses a system called Nitro to offload virtualization, networking and storage management from the server processor onto a custom chip.
Dedicated GPU infrastructure can beat the public cloud on cost. Companies considering purchasing an AI cluster need to consider utilization as the key variable in their calculations.
The number of proposals for new hyperscale-size data centers have reached new heights in 2024. Those that are built will require huge investment and resources — but many proposals will fail to move forward.
To meet the demand driven by AI workloads, a new breed of cloud provider has emerged, delivering inexpensive GPU infrastructure as a service. Their services are highly demanded today, but longer-term, the market is ripe for consolidation.
Although quantum computing promises a revolution in scientific discovery, its use is still constrained to research and continuing development. However, a new IBM quantum data center in Germany signals a growing interest in its capabilities.
Hyperscalers use DC busbars in their racks. Proven at large scale, the technology saves energy and materials, but without a market for DC servers that meets the needs of smaller operators and enterprises, no one else will use it.
The 14th edition of the Uptime Institute Global Data Center Survey highlights the experiences and strategies of vendors, product suppliers, engineering firms, and consultants in the areas of data center spending, customer challenges, supply chains, D...
Gigantic “gigawatt” data center campuses are being proposed or planned in many locations around the world. They could change the digital infrastructure landscape, but many questions — including how many will be built — remain.
Cloud providers advise users to build application resiliency by deploying across multiple locations. This report quantifies the emissions, resiliency and costs of different stateless cloud-application architectures.
To improve the resiliency of an application in a public cloud, it is often distributed across data centers. This report quantifies the costs, levels of resiliency and outages compensation for common public cloud architectures.
Avoiding downtime remains a top priority for managers of critical digital infrastructure. As the demands placed on IT change, so do the types, frequency and impacts of outages. This report draws lessons from recent outage data.
Owen Rogers
John O'Brien
Peter Judge
Anthony Sbarra
Laurie Williams
Andy Lawrence
Douglas Donnellan
Lenny Simon