The US state of Virginia has retained its sales-and-use-tax exemptions for data centers but is imposing a new tax on electricity use: a move that highlights the trade-offs, uncertainties and potential spread of industry-specific taxation.
For years, the industry has celebrated a reassuring trend: despite a growing number of outages, resiliency has continued to improve. But as AI-driven expansion accelerates, that long-running improvement may be about to stall: or even reverse.
A new framework from the Electric Power Research Institute (EPRI) creates a classification system for data center demand response capabilities to help promote grid flexibility and, in turn, support the continued growth of digital infrastructure.
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.
AI data centers are racing ahead, but the grid isn't keeping up: operators need to rethink how they connect to, generate and manage energy to unlock expansion without overwhelming already strained generation and transmission systems.
In AI model training, idle GPUs — not high prices — are the biggest driver of cost, with poor utilization quietly burning tens of thousands of dollars in wasted compute capacity.
An alert from the North American grid connection authority shows that data centers will be treated similarly to generation assets when requesting power connections, requiring operators to share more information and permit operational monitoring.
The 16th edition of the Uptime Institute Global Data Center Survey highlights the experiences and strategies of data center owners and operators in the areas of resiliency, sustainability, efficiency, staffing, cloud and AI. The attached data files…
Policymakers across the US are reassessing tax incentives for data center operators, reflecting shifting priorities around infrastructure, costs and community impacts.
The cost of AI training varies depending on underlying infrastructure and the training approach adopted. This web tool calculates training times and workload infrastructure costs based on configurable data center, infrastructure and model attributes.
Growing public opposition to data center development has pushed many data center companies to invest in media and promotional campaigns. But, as opposition intensifies, this type of outreach is unlikely to sway public opinion.
Although cloud platforms often offer the lowest cost for AI inference, on-premises deployment may be preferable due to application architecture, data locality and control requirements.
The cost of AI inference varies widely depending on deployment model, utilization and hardware. This costing tool compares on-premises, colocation and managed AI platforms on a like-for-like basis.
PUE measures facility energy overhead relative to IT load, but it does not show how much of a site's provisioned power envelope ultimately reaches IT. Power and compute effectiveness (PCE) introduces a metric that makes this allocation visible.
Uptime Intelligence analysis shows that minimum thresholds for IT power utilization misrepresents server work capacity utilization. These thresholds can unintentionally incentivize operators to disable server energy efficiency features.
Douglas Donnellan
Andy Lawrence
Dr. Tomas Rahkonen
Dr. Owen Rogers
Jay Dietrich
Peter Judge
Paul Carton
Anthony Sbarra
Laurie Williams
Rose Weinschenk
Dr. Rand Talib