RSS Amplifier

SEMIVISION @_@ · Aug 17, 2026

OCP APAC 2026: AI Infrastructure Is Moving Beyond the GPU

0
Sign in to vote or save

SEMIVISION · SEMIVISION @_@

The most important takeaway from OCP APAC 2026 is not another GPU roadmap.

It is that the AI infrastructure bottleneck is moving outward.

For the last three years, the AI trade has been dominated by a relatively concentrated set of components: GPUs, HBM, advanced-node wafers, and advanced packaging. Those remain critical, but the next phase of deployment is exposing constraints elsewhere in the system.

  • Power distribution is becoming harder.

  • Cooling is becoming a system-level problem.

  • Fiber density is increasing faster than data-center physical infrastructure was designed to handle.

  • Management complexity is rising.

  • Packages are getting larger.

  • And validation equipment increasingly needs to handle power levels that would have looked absurd for a server only a few years ago.

That is why OCP APAC increasingly looks less like a server conference and more like an AI infrastructure supply-chain conference.

The companies worth watching are therefore not limited to NVIDIA, AMD, Broadcom, or the server ODMs.

The next layer of value creation is moving into power, thermal management, optics, system control, packaging, and test.

SemiVision attended the first OCP APAC Summit in Taiwan last year, and we were already impressed by the scale and quality of the event. This year, it feels even bigger. Last year, the event was mainly held on the 7th floor of Taipei Nangang Exhibition Center Hall 2; this year, the exhibition booths and dining area have expanded to the 4th floor as well.

You can clearly feel that more companies, engineers, and industry leaders are joining the OCP community. Taiwan’s role as a critical hub for AI and semiconductors makes it a natural meeting point for global companies to visit the supply chain, exchange ideas, and build new connections. We look forward to seeing OCP APAC continue to grow bigger and stronger in Taiwan.

Below we will share:

  • AI Infrastructure Is Shifting From Servers to Racks

  • Power Is Becoming One of AI’s Hardest Engineering Problems

  • Liquid Cooling Has Entered the Qualification Phase

  • KAORI and the Heat-Exchange Layer

  • AI Networking Is Running Out of Physical Space

  • CPO Is an Ecosystem Problem

  • AI’s Bottleneck Is Moving Through the Stack

Read the original on tspasemiconductor.substack.com

Comments

Nothing yet. Say the first thing.

    Sign in to join the conversation.