FS launches 800G LPO module for low-latency, high-efficiency data center interconnects

FS has introduced its 800G Linear Pluggable Optics (LPO) module, designed for high-density data centers supporting artificial intelligence and machine learning workloads. The 800G DR8 OSFP finned top LPO module targets hyperscale AI cluster deployments by offering high-speed optical connectivity with reduced power consumption and lower latency compared to traditional digital signal processor (DSP)-based optics, according to FS.

FS reports that its 800G LPO DR8 module operates with maximum power consumption of 8.5 W, which it claims is approximately 50 percent lower than DSP-based 800G modules. This reduction is intended to decrease heat load on switches and servers, lowering cooling demands and operational costs in large-scale data centers.

By removing the DSP chip from the optical module architecture, FS states that its LPO module also reduces end-to-end latency, a benefit for real-time and latency-sensitive applications such as AI model training and supercomputing. The module uses transimpedance amplifier (TIA) and driver chips with high linearity and equalization, which FS says slightly increases component cost but results in a lower total cost of ownership for large deployments.

According to FS, the 800G LPO module has been tested through traffic tests, bit error rate (BER) evaluations, and optical spectrum examinations to verify stability and reliability. Field testing with FS LPO switches demonstrated error-free operation under operational workloads.

“FS is committed to driving next-generation optical innovation,” said Jon Zheng, FS’s senior technical director. “The 800G LPO module reflects our dedication to delivering high-performance, energy-efficient, and reliable connectivity solutions for global clients, empowering them to secure competitive advantages in modern data center deployments.”

The primary applications for the FS 800G LPO module include large-scale data centers focused on artificial intelligence and high-performance computing, as cited by FS. Other mentioned verticals include telecom, enterprise, and service providers.

Source: FS

Get Data Center Engineering News In Your Inbox:

Popular Posts:

logo_on-white
Vultr introduces AMD Instinct MI355X GPUs to enhance AI workload performance and efficiency in data centers
Grid8-graph
Grid8 launches free interconnection dashboard for US data center and energy operators
GmyybkpD3o7f45xka2pctu42nog
FS launches high-density 1U WDM platform for data centers, maximizing rack space and passive cooling
pr-image-forcadence-and-samsung-foundry-expand-collaboration-to-accelerate-ai-data-center-chip-design-and-power-optimization-jpg
Cadence and Samsung Foundry expand collaboration to accelerate AI data center chip design and power optimization
Screenshot
Knowles launches safety-certified high-voltage ceramic capacitors for data center applications

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: