Intel courts developers with open-source tools and novel hardware architectures to compete in AI-accelerated workloads.

Intel is attempting to differentiate itself from competitors by offering developers early access to unreleased hardware and emphasizing an open-source approach. Markus Flierl, corporate vice president for Intel's Developer Cloud and an IT veteran who previously survived Sun's absorption by Oracle and served as head of GPU cloud infrastructure at Nvidia, describes Intel's strategy as "fundamentally different" from Nvidia's more proprietary model. "Everything is open," he says. "And we very much encourage the participation of the community to help work with us, as opposed to the more proprietary strategy that Nvidia is pursuing." Intel recently open sourced the Continuous Profiler, developed by Intel Granulate, enabling development teams to identify bottlenecks in code and optimize applications for maximum efficiency on Intel hardware.

Intel's Developer Cloud aims to give developers access to cutting-edge hardware typically several months before cloud competitors can obtain the silicon. However, Intel currently trails major cloud providers significantly in datacenter capacity. "We have datacenters predominantly in North America right now," Flierl explains. "They are leased – it's just a timing issue. I just came on board two years ago [but] it takes three years to build a datacenter." The company faces additional pressures from surging demand for AI infrastructure. "It's getting increasingly hard for datacenters with all the demand for AI … these are very power-hungry workloads."

To expand its global presence, Intel is diversifying its datacenter locations beyond North America. "It's probably not going to be in North America right now. I'm on the lookout for a datacenter here in Europe, and I'm also looking at APAC," Flierl states. The company is also exploring partnerships with local providers to establish a sovereign cloud offering.

Flierl emphasizes that Intel's Developer Cloud operates as "a two-way street," providing early access to strategic customers while enabling direct feedback from end users. This direct-to-customer model distinguishes Intel from its traditional approach of selling to OEMs and cloud service providers. "We're the only company in the world that fabricates its own chips, designs its own chips, and also makes it available as a cloud service," he notes. While Google builds its own TensorFlow chips, those remain unavailable for purchase elsewhere, giving Intel unique advantages.

The service offers developers substantial flexibility across the technology stack, from bare-metal Xeons to managed Kubernetes clusters. "You can come in at the different layers of the stack, depending on what you're trying to do ... and then those services are available across the different instance types, so you can see – this workload – how well does it run on CPU versus how well it runs on a Gaudi versus how fast it runs on a GPU, and based on the results you're seeing, you're going to optimize your workload," Flierl explains.