Micron's Vertical HBM Targets 2028 Amid Cooling Challenges

What Happened

Micron Technology is developing a next-generation vertical high-bandwidth memory architecture designed to increase AI system performance, with production targeted for 2028, according to a report published by Benzinga. The company has identified thermal management as a central engineering challenge that will influence how and where the technology is adopted.

Background

High-bandwidth memory, known as HBM, has become a critical component in AI infrastructure. It is used alongside graphics processing units and AI accelerators to move large volumes of data at high speed, a requirement that has grown sharply as AI model sizes and training workloads have expanded. Micron is one of three major DRAM manufacturers globally, competing with Samsung Electronics and SK Hynix in the HBM market.

SK Hynix currently holds a leading position in HBM supply, having secured a significant share of orders from Nvidia, whose H100 and successor chips rely on HBM for memory bandwidth. Micron has been working to expand its own HBM production capacity and has previously confirmed HBM3E shipments to major customers.

Vertical HBM refers to an architectural approach that stacks memory dies in a taller configuration than current designs, with the goal of increasing bandwidth density without proportionally increasing the physical footprint. This approach is distinct from conventional HBM stacking methods already in production.

The Cooling Problem

According to the Benzinga report, the vertical HBM architecture introduces thermal challenges that are not fully resolved. Stacking memory dies more densely increases heat generation within a confined area. Existing cooling solutions used in AI server configurations, including air cooling and some liquid cooling setups, may not be sufficient to manage the heat output produced by the new design at scale.

The report indicates that cooling constraints are expected to shape adoption timelines and deployment environments for the technology. Data centers with advanced liquid cooling or immersion cooling infrastructure would be better positioned to integrate vertical HBM chips than facilities relying on conventional thermal management systems.

Micron has not publicly detailed the specific thermal thresholds involved or the cooling specifications it is working toward, based on available information from the wire report.

Market Context

Demand for HBM has accelerated alongside investment in AI training clusters and inference infrastructure. Analysts and industry observers have noted persistent supply tightness for HBM3E in 2024 and into 2025, as chip designers raced to integrate more memory bandwidth into their products. Micron's 2028 production target for vertical HBM places the technology in a timeframe that coincides with what several AI hardware companies have signaled as a next major upgrade cycle.

The cooling challenge cited in connection with Micron's vertical HBM is not unique to this product. The broader AI hardware industry has faced increasing pressure on data center power and thermal management as GPU and accelerator chip power envelopes have risen. Several major hyperscale operators, including Microsoft, Google, and Meta, have publicly discussed investments in advanced cooling infrastructure to support next-generation AI hardware.

What It Means in Practice

If cooling constraints are addressed, vertical HBM could offer AI system designers a path to higher memory bandwidth without requiring a larger chip package footprint. This would be relevant for accelerator designs where physical space and power distribution are limiting factors. If the constraints are not resolved, adoption may be limited to data centers equipped with specialized cooling systems, potentially narrowing the addressable market at launch.

Micron has not issued an official press release or investor guidance specifically addressing the vertical HBM timeline or the cooling challenge as described in the Benzinga report, and the company has not publicly confirmed the details cited.

Micron is scheduled to report its next quarterly earnings in the coming weeks, where analysts are expected to seek further detail on HBM production roadmaps and customer commitments.