The AI industry has spent the past few years trying to give models access to more information. But bigger context windows come at a cost, requiring more memory, more compute and more infrastructure. For organizations building long-running AI systems and agents, managing context is becoming a challenge in its own right. A new research paper…
Code generation got faster while delivery got harder. AI coding tools solved one problem for engineering teams. They made it dramatically easier to generate code. They also put tremendous stress on everything that comes after. Many teams now live in a strange reality where pull requests move faster and code volume keeps climbing, however delivery…
Cloud data giant Snowflake has committed $6B to AWS through 2032, reflecting growing confidence in future demand for its platform. The deal is the latest example of how the biggest winners in the AI era may not be the companies building models, but the companies that enable enterprise data. The agreement extends the two companies’…
The same dataset should not need to exist twice just to be useful. Yet in most environments, it does. One version lives in a file system, shaped for enterprise users and applications that expect paths, directories, and mutable state. Another version is exported into object storage so distributed engines and AI pipelines can process it…
Elon Musk’s SpaceX has filed for what could become the largest IPO in history. It could potentially raise a staggering $75 billion. OpenAI is targeting a public offering as soon as September, based on reports. Anthropic is said to be exploring a listing later this year. No surprise that these IPOs have generated a lot…
AI for science is changing what storage infrastructure needs to do. Most of this infrastructure was originally built around simulations and long term storage. Now AI systems need constant and real time access to massive amounts of scientific data. As a result, the role of storage inside scientific computing environments is starting to change. Many…
The GenAI boom has made hardware hot, both literally and figuratively. Unfortunately, the huge demand for infrastructure has completely disrupted the supply chain for chips, memory, and disk, making it nearly impossible to get the sort of hardware you need to run enterprise IT workloads–let alone HPC or AI jobs–without breaking the bank. So how…
For years, hyperscalers built infrastructure largely for their own cloud ecosystems. However, AI is starting to break that model. The sheer cost of accelerators, data center expansion, and power requirements is pushing AI infrastructure toward a more distributed financing approach, where outside capital plays a much larger role. Google’s reported TPU partnership with Blackstone is…
The reported $66.8B bid by NextEra Energy for Dominion Energy is one of the largest utility mergers in U.S. history. At BigDATAwire we have covered how electricity has emerged as a major bottleneck for AI. Yes, the hardware matters, but those aren’t any good if they don’t get the power they need. Training clusters and…
AI infrastructure and software company Scale AI has signed an MOU with the Department of Energy to support the Genesis Mission – the collaborative AI for science initiative bringing together leading national labs and advanced computing environments. The move adds another commercial AI infrastructure player to an ecosystem that is increasingly centered around integrating AI…
