MWC Barcelona—NTT Corporation and Red Hat, in collaboration with NVIDIA and Fujitsu, have developed a new solution aimed at enhancing real-time artificial intelligence (AI) data analysis at the network edge, said a statement.
The push, part of the Innovative Optical and Wireless Network (IOWN) initiative, leverages the IOWN Global Forum's technologies and Red Hat OpenShift to create a system recognised for its real-world viability.
The solution addresses the growing demand for AI inferencing and the critical role of telco edge in this field. “The demand for AI inferencing is growing, and telco edge has a pivotal role to play,” said Ronnie Vasishta, senior vice president of telecom at NVIDIA.
He further explained the collaboration's focus on combining the APN network with accelerated data processing and AI to deliver low latency and power-efficient computer-vision and image-processing technology.
At the heart of this development is the IOWN All-Photonics Network (APN) and data pipeline acceleration technologies, which enable efficient collection and processing of large sensor data volumes at the edge.
NTT's use of Remote Direct Memory Access (RDMA) over APN for accelerated data pipeline for AI demonstrates significant reductions in power consumption and latency, making real-time AI analysis more feasible and efficient.
The proof of concept (PoC) for this solution involved testing a real-time AI analysis platform with sensor installations in Yokosuka City and a remote data center in Musashino City, connected via APN.
According to the Japanese firm, the PoC showed a 60% reduction in latency and a 40% reduction in power consumption per camera for AI analysis at the edge, compared to conventional methods.
Chris Wright, CTO and senior vice president of Global Engineering at Red Hat, asserted the significance of this work within the IOWN Global Forum, stating, “This is important and exciting work, and these results help prove that we can build AI-enabled solutions that are sustainable and innovative for businesses across the globe.”
Katsuhiko Kawazoe, senior executive vice president of NTT and chairman of IOWN Global Forum, said that the solution not only promises reduced data collection overhead and enhanced AI analysis speed but also supports the utilisation of renewable energy sources and increased security through video camera sensors.
