ENET, part of the Network Infrastructure division of NSI Industries, introduced a new 1.6Tbps DR8 OSFP224 optical transceiver designed for AI, high-performance computing (HPC), and hyperscale data center environments. The module targets next-generation switching and accelerator fabrics that rely on 224G electrical signaling to scale bandwidth for large GPU clusters and AI workloads.
The ENET 1.6T DR8 OSFP224 delivers an aggregate throughput of 1.6Tbps across eight 200G PAM4 electrical lanes. The module uses a DR8 parallel single-mode fiber interface and supports both Ethernet and InfiniBand protocols. The OSFP224 form factor aligns with emerging switch ASIC and accelerator platforms supporting 224G electrical interfaces, enabling higher switch capacities, denser port configurations, and improved power efficiency in modern AI fabrics.
ENET said the module targets spine, leaf, and accelerator interconnect topologies used in AI clusters. Integrated digital diagnostics using CMIS 5.3 provide real-time monitoring of optical power, temperature, voltage, and other operational parameters, helping operators manage and maintain large-scale deployments. The 1.6T DR8 OSFP224 is currently available for qualification and deployment.
• 1.6Tbps aggregate data rate
• 8 × 200G PAM4 electrical lanes
• DR8 parallel single-mode fiber interface
• OSFP224 form factor supporting 224G electrical signaling
• Supports Ethernet or InfiniBand protocols
• Designed for AI, HPC, and hyperscale data center fabrics
“As AI models grow larger and clusters become more complex, networks need to offer consistent, high-speed performance,” said Jason Barrette, VP of Sales and Operations at ENET. “Our 1.6T DR8 OSFP224 is designed for demanding GPU interconnects and advanced switching environments, where bandwidth, density, and reliability are crucial for maximizing computing power.”
🌐 Analysis: The emergence of 1.6T optical modules reflects the industry’s shift toward 224G electrical signaling and 200G per-lane optical interfaces as hyperscale operators scale AI clusters. These optics will align with upcoming 51.2Tbps and 102.4Tbps switch ASIC generations designed to support high-radix AI fabrics connecting thousands of GPUs.
🌐 We’re tracking the latest developments in networking silicon. Follow our ongoing coverage at: https://convergedigest.com/category/semiconductors/
🌐 We’re launching the “Data Center Networking for AI” series on NextGenInfra.io and inviting companies building real solutions—silicon, optics, fabrics, switches, software, orchestration—to share their views on video and in our expert report. To get involved, send a note to [email protected] or [email protected].
