Huawei Technologies has unveiled a high-speed optical module intended to remove bottlenecks in artificial intelligence infrastructure, showcasing what it called the industry’s first 7.2-terabit-per-second (Tbps) near-packaged optics (NPO) product at the China International Optoelectronic Exposition (CIOE) in Shenzhen this week.
Optical module design and how it addresses AI bottlenecks
The Shenzhen-based company said the technology replaces traditional copper wiring with light-based signals to move large volumes of data between AI processors at higher speeds and with lower power consumption. As AI clusters expand to thousands of interconnected processors, Huawei said conventional copper connections are reaching physical limits in bandwidth, transmission distance and power efficiency; NPO brings optical components closer to network switching chips, shortening electrical connections and reducing signal loss and power use.
Huawei’s module combines 36 channels operating at 200 gigabits per second each, giving an aggregate bandwidth of 7.2Tbps, according to Man Jiangwei, director of Huawei’s advanced optoelectronics laboratory. That capacity places the product among a new generation of high-density optical interconnects for AI systems, alongside 6.4Tbps optical engines developed for co-packaged optics by US chipmaker Broadcom.
“We have already fully entered the product development stage,” Man Jiangwei said in an interview on Thursday. “Once the whole supply chain and manufacturing system is in place, I believe it won’t be long before we can move quickly into volume production.”
