Micron Starts Mass Production Of High-Speed Memory Semiconductors For Nvidia AI

Micron Technology has started mass production of high-speed memory semiconductors (photo: x @MicronTech)

JAKARTA - Micron Technology has started mass production of high-speed memory semiconductors for use in Nvidia's latest chips for artificial intelligence. This has made its shares rise more than 4% before the bell on Monday, February 26.

According to Micron, the HBM3E (High Bandwidth Memory 3E) will consume 30% less power compared to competing offerings and can help capitalize on the rapidly increasing demand for chips that power a generating AI app.

Nvidia will use the chip in its next-generation H200 graphic processing unit, which is expected to start shipping in the second quarter and replace the current H100 chip that has pushed for a massive revenue spike in chip designers.

Demand for high-speed memory chips (HBM), a market led by supplier Nvidia SK Hynix, for use in AI has also raised investor expectations that Micron will be able to survive a slow recovery in other markets.

HBM is one of the most profitable products for Micron, partly because of the technical complexity involved in its construction.

The company had previously said it expects "a few hundred million" HBM revenue dollars in the 2024 fiscal year and sustainable growth by 2025.

HBM3E is claimed to consume 30% less power compared to its competing offerings. This means the use of HBM3E can produce higher power efficiency, which is important for applications that require high performance such as artificial intelligence (AI) generative.

Despite consuming less power, HBM3E offers high performance in transferring data with a large bandwidth. This allows chips that use HBM3E to work efficiently in dealing with computationally intensive workloads, such as complex graphic rendering in AI applications.

HBM3E is the latest high-speed memory technology, with the latest features and improved performance compared to previous versions. This makes it an attractive choice for companies like Nvidia to use in their latest products that require high performance and good power efficiency.