Intel and Google recently announced a collaboration on the use of Intel Xeon processors for Google’s data center infrastructure and a collaboration on the development of a new class of chips called “Infrastructure Processing Units” or IPUs.
With the signing of this collaboration, Google Cloud will use Intel Xeon 6 processors to power a variety of Google servers, including the C4 and N4 servers responsible for training artificial intelligence systems, inference processing and also general applications such as cloud storage and media storage.
In addition, Google Cloud will also use Intel Xeon processors from different generations to reduce data processing capacity and energy consumption, where not all Google servers require the latest and most powerful processing chips from Intel.
Intel and Google are also expanding the joint development of their custom ASIC-based IPUs. These programmable accelerators are expected to offload network, storage and security functions from CPU components. improve utilization, increase efficiency and enable more predictable performance across AI environments.
IPUs are now seen as a critical component in modern data center architectures, handling infrastructure tasks traditionally handled by CPUs. This component unlocks more efficient computing capacity and allows cloud providers to scale more efficiently without increasing overall system complexity.
The combination of Xeon CPUs and IPUs forms an integrated platform that balances general-purpose computing with purpose-built infrastructure acceleration to deliver more efficient, flexible, and scalable AI systems.
