Marvell PCIe 6 retimers adopted by leading data center server and AI infrastructure providers

Marvell Technology has announced that major server manufacturers and cable suppliers are deploying its Alaska P PCIe 6 retimer product line to boost connectivity and performance in advanced data center systems, particularly those supporting accelerated artificial intelligence (AI) infrastructure. The company reports that these retimers are enabling high-speed, low-latency, and low-power connections between AI accelerators, general-purpose processors, GPU and XPU platforms, storage devices, Compute Express Link (CXL) components, and other crucial data center elements.

According to Marvell, server vendors have adopted Alaska P PCIe 6 retimers for GPU and XPU platforms and as add-in cards for general-purpose servers. The technology is also available in PCIe active electrical cables and active optical cables supplied by various cable and optical partners, with ongoing evaluation for integration into storage systems to enhance signal quality between CPUs and solid-state drives (SSDs).

Marvell states that its PCIe 6 retimers are based on 5 nanometer (nm) Pulse Amplitude Modulation 4-level (PAM4) serializer/deserializer (SerDes) technology. The devices operate at 64 gigatransfers per second (GT/s), supporting compensation for up to 40 dB of channel loss while offering advanced telemetry, diagnostics, and fleet-management features. These specifications are intended to maintain signal integrity and enable large-scale, low-latency interconnects in accelerator-centric server and cluster architectures, which Marvell identifies as critical for managing increasing data flows in data center AI and machine learning workloads.

The Alaska P PCIe retimers are designed for flexible data center architectures. They can be deployed on server boards, within copper cable assemblies, or integrated with electrical-to-optical components for optical PCIe connections. Marvell claims this versatility allows cloud operators to scale disaggregated compute fabrics in AI data centers efficiently.

Vendor and partner perspectives were included. Xi Wang, senior vice president and general manager of the Connectivity Business Unit at Marvell, stated, “The adoption of the Alaska P PCIe retimers underscores our leadership in enabling the AI infrastructure transition from traditional server architectures to disaggregated, accelerator-centric compute fabrics,” adding that “Our SerDes innovation and deep ecosystem collaboration are helping the world’s top OEMs and interconnect partners scale performance for demanding AI workloads.”

Marvell reports that Alaska P PCIe retimers are available now through system vendors and cable partners.

Source: Marvell Technology

Get Data Center Engineering News In Your Inbox:

Popular Posts:

1765906506220
Tritium launches 800 VDC bidirectional inverter for data centers and renewable energy sites
Water_From_Air_Data_Center_Rendering_Nov_2025
iMasons selects Water From Air for data center waste heat-to-water innovation
Screenshot
Five AI data centers to reach 1 GW power capacity in 2026, new analysis shows
68e79d30a17eea847251fae6_img-home-product-liquidjet-main
Frore Systems updates LiquidJet direct-to-chip coldplate for 1,950 W NVIDIA Rubin data center GPUs
River Bend Rendering
Hut 8 announces $7 billion, 245 MW hyperscale data center IT lease with Fluidstack backed by Google

Share Your Data Center Engineering News

Do you have a new product announcement, webinar, whitepaper, or article topic? 

Get Data Center Engineering News In Your Inbox: