Scaling AI compute requires more than improving accelerator performance. It increasingly depends on how efficiently data moves between accelerators, memory, CPUs, and supporting resources. This need is driving major changes across compute, memory, and interconnect architectures.

Join us for the first webinar in our three-part AI Scale-Up Insider Series.

In Exploring the Technologies Driving AI Compute Forward, we'll examine the architectural trends reshaping AI infrastructure. You’ll learn how evolving technologies such as PCI Express® (PCIe®), UALink, and advanced memory address bottlenecks in bandwidth, latency, coherency, and accelerator connectivity. We’ll also explore why higher data rates, denser interfaces, and tighter margins make interoperability, signal integrity, and compliance validation increasingly difficult.

You’ll learn:

• How modern AI scale-up architectures move data between compute and memory.

• The roles of compute I/O, accelerator fabrics, and advanced memory standards.

• Why next-generation interfaces are creating new interoperability and validation challenges.

Register today to gain a clear understanding of the technologies driving AI infrastructure scale-up and the validation challenges they create.

PCIe® and PCI Express® are US registered trademarks and/or service marks of PCI-SIG.