In low bandwidth, high latency environments the ability to reduce and prioritize information can have a dramatic effect on the overall network health and optimization. SliceUp’s edge database and OpsIQ module provides inference based cognitive Lambda, downsampling, parameterization, anomaly detection, and a context aware QOS classifier. These capabilities allow our customers to gain real-time anomaly detection, significantly reduce its bandwidth utilization and take immediate, automated action.
When dealing with network communications, there is an inherent struggle between the need for more information and the conservation of data bandwidth. In order to increase observability, the more data generated by endpoints and sensors the better visibility that users have into how well all systems are functioning. The nature of the network demands that choices are made in order to reduce network consumption from observability parameters. However, those decisions typically come at a steep price for operations.
Our context aware QOS classifier optimizes network bandwidth to ensure high-priority information is immediately available. Using anomaly detection directly at the edge, we identify the information that requires immediate attention and transmit that first, via the path with least latency. Then, as bandwidth availability permits, the nominal data is downsampled and sent using cognitive lamba technology, where less important information is sent to low cost storage for completeness or removed if it's not needed.