Marvell Advances Interconnect Portfolio for Scale-up and Scale-out Fabrics at OFC 2025
Comprehensive Technology and Strong Ecosystem Support Delivering Higher Performance, Customization and Flexibility for Rack-, Row- and Cloud-scale AI Networks
SANTA CLARA, Calif., March 31, 2025 — Marvell Technology, Inc. (NASDAQ: MRVL), a leader in data infrastructure semiconductor solutions, today announced it will showcase its interconnect portfolio for scale-up and scale-out AI deployments at OFC 2025, taking place March 31–April 3 in San Francisco, CA. Marvell will share its strategy and vision through multiple presentations and demonstrations in its booth #2129 and partner booths.
Interconnect technology—the semiconductors, software and system that serve as a nervous system for reliably and rapidly transmitting data across accelerated infrastructure—is in the midst of a revolution. The rapid buildout of data infrastructure worldwide is expected to more than triple the market for interconnect chipsets to $11.5 billion by 2030, according to LightCounting. Additionally, interconnect technology will undergo a fundamental transformation, advancing to support higher speeds, extended distances, reduced latencies and increased data traffic volumes essential for training, inference and other AI and cloud services in the coming years.
For scale-up fabrics in racks and rows, co-packaged optics (CPO), linear packaged optics (LPO), co-packaged copper (CPC), active electrical cable (AEC) and PCIe retimer interconnect technologies will pave the way for row-scale computing systems containing hundreds of XPUs and CPUs several meters apart spread over multiple racks that can outperform today’s systems in performance-per-watt and time-to-completion. New scale-out networking technologies such as 400G/lane networking and coherent-lite DSPs will greatly increase the carrying capacity of networks while extending the reach of links from meters to kilometers for more powerful, versatile cloud-scale infrastructure.
Products, technology and ecosystem initiatives Marvell will showcase at OFC 2025 include:
- 400G PAM4 Technology. A live demonstration of the industry’s first 400G/lane technology with a complete electrical to optical link operating at 224 Gbaud. 400G is a critical step towards 3.2T optical interconnects and 204.8T switches.
- Co-Packaged Platforms for AI Scale Up and Scale Out. Marvell to showcase its CPO and CPC technologies for achieving higher interconnect densities and longer reach. Marvell will showcase system-level XPU and switch implementations supporting rapid servicing and ease of manufacture.
- 1.6T Silicon Photonics Light Engine for Scale-up Networks. Marvell will show its new 1.6T silicon photonics light engine integrated into a 1.6T LPO reference design module operating at 200G/lane.
- 200G/Lambda 1.6T PAM4 Optical Interconnect for AI Scale Out. Marvell will demonstrate Ara, the industry’s first 3nm 1.6T PAM4 interconnect platform featuring 200 Gbps electrical and optical interfaces.
- 800G ZR/ZR+ Pluggable Optics for Multi-Site AI Training. A live demonstration of COLORZ® 800 pluggable DCI module operating in several modes, including 800G ZR and 800G ZR+ with probabilistic constellation shaping, allowing for 800 Gbps transmission over distances of up to 1,000km.
- 200G/Lane 1.6T AEC for AI Scale Up and Scale Out. A live demonstration showing production-ready Alaska® A AEC DSPs operating at 8 x 200G delivering 1.6T total bandwidth for multi-rack scale-up systems. In addition, multiple partners will showcase in their booths Alaska A 1.6T AEC DSPs for emerging 200G/lane-based accelerated infrastructure and Alaska A 800G AEC DSPs that deliver connectivity for up to seven meters.
- PCIe Gen 6 and PCIe Gen 7 SerDes End-to-End Over Optics. A hands-on demonstration of an Alaska P PCIe Gen 6 retimer driving PCIe signals between the root complex and end point over an optical fiber, using a 100G per lane LPO module supplied by TeraHop. Second technology demonstration highlights the performance of a 128G SerDes circuit designed for integration into future PCIe Gen 7 retimer devices using a TeraHop 200G per lane LPO module.
- 51.2T Scale-out Fabric: Showcasing the switching and interconnect capabilities of its accelerated infrastructure portfolio for AI clusters, this demonstration utilizes CPU-powered servers as compute elements to show data traffic traversing a model AI cluster. The demonstration includes RDMA-enabled network interface cards, copper and optical interconnects, including 7m active electrical cables, and SONiC-enabled switches representing both frontend top-of-rack, middle-of-row, end-of-row leaf and spine applications and backend switch fabrics.
Marvell Presentations and Panels at OFC 2025
Throughout the conference, Marvell executives and technical experts will participate in key presentations, panels and workshops, discussing critical industry topics. The full agenda is available at https://www.marvell.com/company/events/ofc-2025.html.
关于 Marvell
为了通过数据基础设施技术实现全球互联的愿景，我们正在构建非常强大的基础解决方案： 我们与客户携手。 二十五年来，Marvell 深受世界位居前列的技术公司的信赖，以客户当前需求和未来理想为设计路线，运用专业半导体解决方案实现对全球数据的移动、存储、处理与保护。 通过深度协作、践行透明化的沟通方式，我们致力于促进未来企业、云、汽车和运营商架构的不断转型与完善。
# # #
Marvell 和 M 徽标是 Marvell 或其附属公司的商标。请访问 www.marvell.com 网站获取 Marvell 商标的完整列表。其他名称与品牌可能是其他方财产。
本新闻稿包含联邦证券法界定的前瞻性声明，该等声明涉及风险和不确定性。 前瞻性声明包括但不限于任何可能预测、预估、间接提及或暗示未来事件、结果或成就的声明。 实际事件、结果或成就可能与本新闻稿的预期存在重大差异。 前瞻性声明仅为预测，且受难以预测的风险、不确定性和假设的影响，这些风险、不确定性和假设包括 Marvell 的 10-K 年度报表中的“风险因素”部分、10-Q 季度报表以及 Marvell 不时向美国证券交易委员会 (Securities and Exchange Commission, SEC) 提交的其他文件中提及的内容。前瞻性声明仅在发表当日有效。 谨提醒读者不要过分依赖前瞻性声明，无论是由于新信息、未来事件还是其他原因，任何人都没有义务更新或修改任何此类前瞻性声明。
For further information, contact:
pr@marvell.com
