Home 9 Semiconductors 9 Empower, Marvell Partner on Integrated Power Solutions

Empower, Marvell Partner on Integrated Power Solutions

by | Jun 20, 2025

Locating voltage regulation under or near the processor improves power efficiency and system performance. This approach reduces distribution losses and supports the increasing current demands of modern compute architectures.

SAN JOSE, CA, June 20, 2025 – Empower Semiconductor has announced a collaboration with Marvell Technology to develop integrated power solutions for Marvell custom silicon platforms. The advanced solutions are designed to accelerate the transformation of power delivery systems to smaller silicon chips coupled with the processor.

The joint solutions are part of Empower’s mission to address the power delivery challenges of the kilowatt-chip era. By integrating power delivery with advanced processors, Empower and Marvell enable hyperscalers and infrastructure providers to maximize their performance, efficiency, and return on investment (ROI) of artificial intelligence (AI) and cloud data centers.

“The collaboration with Marvell validates Empower’s technology as a foundational enabler of next-generation AI and cloud infrastructure,” commented Tim Phillips, CEO of Empower Semiconductor. “With integrated voltage regulation, we’re delivering power where it’s needed – right at the point of load – with exceptional density, precision and efficiency.”

“Placing integrated voltage regulators (IVR) under, near, or inside the package delivers greater performance and greater efficiency,” said Will Chu, senior vice president and general manager of custom cloud solutions at Marvell. “We look forward to working with Empower to bring pre-validated packaged IVR power solutions to our custom XPU customers.”

The partnership uses Empower’s FinFast technology and a vertical power delivery setup to give system designers tested, compact power solutions. These shift voltage regulation away from the traditional board-based designs, placing it on or near the chip. By putting power delivery closer to the processor, these systems help lower power loss, improve efficiency, and handle the increasing current needs of new XPUs.

Source: Empower Semiconductor

About Empower Semiconductor

Empower Semiconductor, founded in 2014, is a private company headquartered in San Jose, CA, specializing in power management solutions within the semiconductor industry. The company offers integrated voltage regulators (IVRs) and silicon capacitors designed to enhance power efficiency and density in electronic devices. The products serve sectors requiring advanced power management, including data centers, mobile devices, communications, and AI applications. Empower’s technology aims to reduce the energy footprint and total cost of ownership for data centers, thereby powering the AI revolution. As of 2024, Empower Semiconductor employs approximately 52 individuals. The company has raised $76.81 million in funding over multiple investment rounds, reflecting investor confidence in its innovative power management solutions. Empower’s power density and efficiency advancements position it as a key player in addressing the increasing power demands of modern electronic systems.

About Marvell Technology

Marvell Technology, established in 1995, is based in Santa Clara, CA. The company develops semiconductors used in data infrastructure, including ASICs, data processing units (DPUs), storage controllers, and connectivity solutions. Its products are deployed across sectors such as data centers, 5G networks, cloud services, automotive, and enterprise systems. In fiscal year 2024, Marvell reported annual revenue of approx. $6.0 billion. The company operates globally, with key markets in China, the United States, and Southeast Asia. Marvell has expanded its capabilities through acquisitions, including Cavium, Aquantia, and Inphi, which added technologies in areas like AI acceleration, high-speed data transfer, and silicon photonics. With more than 7,000 employees and over 10,000 patents, Marvell focuses on delivering components supporting high-performance compute, storage, and networking applications, including those that enable AI workloads and infrastructure scaling.