Data is created and required in unusual and challenging locations. Locations such as the International Space Station, connected vehicles, manufacturing floors, ships at sea, or the local pharmacy. Although data has historically been hosted in the central data center or cloud, many crucial choices must now be made at the edge.
To reach the human level, the responses and decisions made by machines and applications need to operate in milliseconds or even microseconds. Waiting for more than 100 ms by sending data to the cloud could be the difference between a good or a poor customer experience.
When analysing user data at the edge, such as that from Internet of Things (IoT) devices or industrial machines, the problem becomes dealing with massive amounts of data. Every device generates vast amounts of data every second, which is only useful if it can be stored or analysed. This data becomes much more useful when used locally and subsequently moved to the open hybrid cloud’s core data center.
This solution pattern provides an architecture solution for scenarios in which edge devices generate image data, which must be collected, processed, and stored at the near edge before being utilized to train AI/ML models at the core data center or cloud.