Cignal AI Forecasts Coherent Pluggable Optics to Lead Telecom Bandwidth, with 800ZR+ Driving ‘Scale-Across’ AI Data Center Applications

Cignal AI USA

Overview

Cignal AI’s new Pluggable Applications Forecast reveals that coherent pluggable optics are leading the telecom market in deployed bandwidth, primarily driven by AI data center interconnect demands. Specifically, 800ZR+ optical technology is widely deployed for ‘scale-across’ applications, extending distributed training across multi-site GPU clusters. The forecast provides a detailed analysis of various coherent revenue models, customer segments, and module types, underscoring the critical role of optical communication in the AI era.

In Depth

Key Findings

Cignal AI has released a pivotal Pluggable Applications Forecast, indicating that coherent pluggable optics are now leading the telecom market in deployed bandwidth, primarily fueled by the accelerating demands of AI data center interconnects. The report specifically highlights the widespread adoption of 800ZR+ optical technology for ‘scale-across’ applications, which enable the expansion of distributed training across multi-site GPU clusters. This trend underscores how the exponential growth in AI workloads is making high-bandwidth, low-latency optical communication solutions indispensable for modern computing infrastructures.

Technical / Clinical Details

Coherent pluggable optical modules integrate advanced Digital Signal Processing (DSP) techniques and coherent modulation schemes into compact form factors. The 800ZR+ represents the latest evolution of such technology, facilitating longer-reach and higher-capacity data transmission over existing data center fiber infrastructure. This technology is critical for efficiently connecting GPU clusters across multiple data centers or geographically dispersed facilities, enabling them to function as a single, logical ‘AI factory.’ This approach maximizes the utilization of computational resources for AI training and supports the rapid development and deployment of a wide range of AI models by addressing the latency and bandwidth needs of such distributed systems.

Background & Context

As AI models become increasingly complex and datasets grow to colossal sizes, the computational resources required for AI training can no longer be contained within a single data center. Consequently, ‘scale-across’ networking, which seamlessly connects and orchestrates GPU clusters spanning multiple sites, has become paramount. Cignal AI’s forecast validates the dominance of coherent pluggable optics in this domain, clarifying why NVIDIA and other hyperscalers are rapidly accelerating their investments in optical infrastructure. Notably, NVIDIA CEO Jensen Huang’s proposed ‘copper-and-optics roadmap’ also emphasizes the expanding role of optics in both long-haul and intra-rack connectivity within data centers.

Strategic Significance & Outlook

Cignal AI’s forecast suggests that coherent pluggable optics will play a central role in both AI data centers and the broader telecom market for the foreseeable future. High-capacity, long-reach pluggable modules like 800ZR+ are becoming determining factors in the scalability and efficiency of AI infrastructure. The proliferation of this technology is expected to create significant business opportunities for optical component manufacturers, DSP providers, and data center infrastructure vendors. Looking ahead, the development of even higher-density and lower-power coherent optical solutions is anticipated, which will enable further innovations in AI applications. Investors and industry stakeholders should closely monitor technological advancements and market growth in this critical sector.