Arc flash risks in data centers using 800V DC strongly depend on architecture — a research paper warns operators that they will need to update arc flash safety procedures to handle DC systems.
Giant data center power capacity and investment are at unprecedented levels, with power proposed in the first half of 2026 tracking at 2025 levels.
Redundant power and cooling equipment is expected to reduce outages — but by how much? Survey data shows that operators using 2N configurations were up to 64% less likely to report an outage than those using N.
AI data center projects often run into delays when developers encounter unfamiliar technologies and unexpected technical challenges. Taking a structured, methodical approach can help keep projects on track.
Meta's decision to give away its latest large language model challenges the narrative around AI infrastructure.
Uptime Institute's Annual Global Data Center Survey, now in its 16th year, is the most comprehensive and longest-running study of its kind. The findings in this report highlight the practices and experiences of data center owners and operators in…
AI-accelerated digital twins are central to many AI factory designs and builds. But despite progress in tackling specific power, cooling and IT challenges, most applications remain unproven beyond initial pilots and lack a clear return on investment.
Douglas Donnellan
Andy Lawrence
Daniel Bizo
Dr. Owen Rogers
Peter Judge
John O'Brien
Jacqueline Davis
Max Smolaks
Rose Weinschenk
The Uptime Institute Global Data Center Survey, now in its 16th year, highlights the practices, perceptions, intentions and experiences of data center owners and operators across resiliency, sustainability, efficiency, staffing, cloud and AI.
Data center waste heat does not significantly raise local temperatures, but concerns over the heat island effect demand a robust response that addresses monitoring and mitigation measures.
Results from Uptime Institute's 2026 Data Center Operations and AI Survey (n=867) focus AI training and AI inference workloads hosted in data centers, as well as how AI workloads are being deployed in terms of data center cooling systems and rack…
In this webinar, we outlined how to build a robust operations and management strategy for AI data centers in a time where extreme power densities, liquid cooling complexity, and rapid hardware refresh cycles demanded a fundamentally different…
Internal efforts to maximize token use, combined with changes in LLM pricing structures, have rapidly increased enterprise AI spending. Yet there is often limited visibility into whether such expenditures are creating value.
More data center operators are performing AI training in 2026, and this type of workload continues to be distributed across a wide range of data center venues. No single hosting model has emerged as the default.
This tool — to accompany the Uptime report "The problem with energy per token" — explores how energy and carbon per token vary depending on throughput, utilization, power consumption and grid carbon intensity, even when using the same hardware.
The cost of AI training depends on venue choice and infrastructure utilization. This costing tool — to accompany the Uptime report "Where to deploy AI training" — calculates workload infrastructure costs for colo, cloud and enterprise data centers.
Chris Brown
Paul Carton
Anthony Sbarra
Laurie Williams
Madeleine Kudritzki