Timothy Prickett Morgan

As you might expect, the Hugging Face open weight AI model distribution hub, which has over 3 million AI models, over 1 million datasets, and over 1.4 million interactive spaces that it hosts, was something that AMD had to focus on as it was trying to increase adoption of its Instinct GPU accelerators. But now

It may turn out that HBM stacked DRAM is the most profitable product in the six decade history of commercial information technology. IBM's mainframe engines in the 1970s might be in the same zone, depending on how you allocate costs and revenues. But my gut says in terms of driving revenues and profits, HBM stacked DRAM packaging beats the monolithic systems technology (MST) packaging of the System/370s in the 1970s and the thermal conduction module (TCM) packaging of the System/3081 in the 1980s.

Trying to figure out how much money is sloshing around for AI infrastructure and services is becoming a sport. We are living through the most dramatic IT investment cycle in the history of information technology, rivaling the debut of commercial computing systems six decades ago that made IBM and several other companies fabulously wealthy (relatively speaking) at that time.

For as long as most of us have been watching high-end commercial and technical computing, the compute part of the system – the CPU for five decades then also GPUs for the past decade and a half – has been, without a doubt, the focal point of the system. Memory, storage, and I/O were always important, but they were secondary.

Every CPU maker always had its own and very proprietary way to link multiple CPUs into shared memory clusters and for the most part, this remains true despite the efforts of the CCIX consortium and even the CXL consortium to create a kind of universal glue to make many CPUs look like one much bigger CPU.

Yup, it is that time of the year again. The annual IBM i Marketplace Survey from Fortra is open, and we are asking you to spend some time and give us all some insight into what is going on inside of your IT organization.

Companies are focused on generative AI and how it might be integrated into their applications and their business, but everyone realizes by now that adding GenAI to the systems stack often means needing to upgrade existing back office (transactions) and front office (data analytics) systems so they can feed the AI beast faster and more efficiently. Because there is nothing in the world more expensive than a GPU or an XPU sitting idle when it could be chewing on or spitting out tokens.

If everyone has an AI inference problem, with tokens still being too expensive to generate even with state of the art GPU and XPU accelerators, Microsoft probably has the biggest inference problem in the world.

No matter how much money anyone throws at it, the GenAI boom can only continue to grow as fast as datacenters are built and powered up to accept machinery.

With over 7,000 Power Systems customers, Fortra is one of the largest software conglomerates serving the IBM i and AIX communities and it is by far the largest supplier of systems software for Power Systems. The company is also a key development partner of Big Blue, having nine years ago taken over enhancements to Rational