Autonomous AI Edge Platform

The bursty nature of AI means user demand will be vastly different from typical workloads that 5G networks were built for. This making them a perfect fit for edge computing, but also presents unique challenges for orchestration and management.

Sending workload requests to the network edge removes the inefficiencies of sending them all the way to the core, which would mean increased latency and response times, adversely affecting the user experience. This pages shows the user interface for an autonomous edge orchestration platform, which has been designed to address the aforementioned challenges.

• The request information button sends an AI inference request to the edge platform.
• This is assigned to a specific edge cluster.
• A specific workdload is deployed.
• The deployment is managed by Kubernetes.
• On completion data is returned.
• This can also be cancelled on demand.