Urgent.News

What's breaking now, across thousands of outlets.

AI

Driven by AI boom, chipmakers announce $115 billion merger

Driven by AI boom, chipmakers announce $115 billion merger

Marvell Technology and Lumentum Holdings have agreed to a $115 billion stock swap, marking the largest merger between publicly held tech companies to date. The deal reflects the growing importance of optical networking in the development of data centers for artificial intelligence. Marvell, a leading provider of chips for networking and AI, has been expanding rapidly through acquisitions since CEO Matthew Murphy took the helm in 2016.

Lumentum specializes in lasers and other chips that transmit data signals using light, known as optical networking, which is becoming more prevalent in data centers. The companies argue that the need for AI computing power has far exceeded the capabilities of individual chips, leading to the development of special-purpose AI processors sold by companies like Nvidia.

These processors are designed to work together as a single system and are increasingly connected to form a cohesive network. To facilitate this transition, companies are replacing copper connections with faster laser-based links near individual AI chips, a trend called co-packaged optics. Marvell's CEO Murphy emphasized that the company that successfully navigates this transition will set standards for decades to come.

Nvidia, a major player in networking technology, has invested in both Lumentum and Coherent, as well as Marvell. Other major networking companies include Broadcom, which manufactures networking chips and aids in AI accelerator design. As of the reporting, Lumentum shareholders would receive 4.5 shares of Marvell stock for each share they hold, suggesting that Lumentum investors would own approximately one-third of the combined company.

The transaction has sparked debate, with analysts warning of potential antitrust scrutiny that could delay the deal by up to 18 months.

Written by urgent.news from The Indian Express's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at indianexpress.com →

More in AI

ML.NET inference pipelines in .NET Aspire with Redis caching

Quick Answer ML.NET inference pipelines in .NET Aspire: By tuning plumbing—static inference sessions, Redis caching, gRPC streaming, batching, and KEDA scaling—ML.NET pipelines in .NET Aspire stay…

  • ML.NET inference pipelines achieve under 50 ms p99 latency
  • Redis KV-cache improves cross-instance state sharing
  • Local ONNX + Redis KV-cache recommended for latency and cost control

More from Tuesday 3 November →