Showing 141–160 of 210 articles in data-centres
Amazon’s $200 billion AI push reflects a shift to building ahead of demand while raising new questions about utilization and the pace at which enterprises will follow.
Nvidia says AI infrastructure should be evaluated on cost per token rather than traditional compute metrics, but analysts say the approach may favor hyperscale environments and remain premature for enterprise IT.
Community protests against data center construction are growing rapidly, halting billions in projects and reshaping industry strategies.
As AI workloads scale, Equinix is betting that network operations must become autonomous to keep pace with increasingly dynamic infrastructure demands.
Meta doubles down on custom AI chips with Broadcom, targeting inference efficiency and Ethernet-scaled infrastructure as workloads grow.
Grid queues, community pressures, and AI demand push operators toward behind-the-meter generation, microgrids, and flexible power.
Industry leaders analyze the proposed pause on AI data centers and the need to balance growth, sustainability, and community concerns.
With demand outpacing capacity, the partners aim to cut timelines and execution risk by unifying planning, construction, and commissioning under a single platform.
As AI buildouts accelerate, strain on materials for servers, networking, and power systems is becoming more evident.
Key initiatives include an addendum to ANSI/TIA-942 for AI infrastructure, the DCE 9000 quality management standard for supply chains, and expanded global certifications.
UK data centers are critical to digital infrastructure but face growing regulatory scrutiny on privacy, cybersecurity, and compliance.
Texas and the Midwest are set to capture more than half of new US hyperscale capacity as AI demand pushes operators toward power-rich regions.
The company advocates for workload-specific memory architectures, such as LPDDR5X, to optimize energy efficiency and performance, signaling a shift away from traditional one-size-fits-all server memory designs.
The company’s new managed agents aim to remove infrastructure bottlenecks, shifting control of complex AI workloads into its platform as enterprises push toward production.
As capital markets tighten and new financing models emerge, enterprises are confronting a key constraint: much of the existing data center footprint was not designed for production AI.
Discover how smartphones could revolutionize edge computing, AI, and sustainability in today’s resource-strained data center world.
Intel and Google are expanding their partnership to prioritize CPUs and IPUs, addressing the growing need for system-level efficiency in AI infrastructure.
Data center growth is colliding with fragmented permitting, local resistance and rising litigation, leaving developers and regulators on uncertain ground.
Short-term GPU bridge loans and tokenization highlight growing strain in AI funding, with just 22.8% of enterprise AI projects meeting ROI goals.
Microsoft’s push into dedicated power with Chevron underscores how competition for AI infrastructure is shifting from chips to electricity.