How to deliver growing data from the end user to the DC?

The growth of peripheral computing is one of the major challenges that many organizations will face in the coming years. Cisco predicts that by 2021 there will be 23 billion connected devices, and Gartner and IDC predict 20.8 billion and 28.1 billion, respectively, by 2020. The role of the periphery will be to ensure the effective use of data from the connected devices, and in their delivery to data centers.

Probably the most widely discussed example of a peripheral computing application with a large amount of data transfer is the delivery of high definition content. In 2016, video accounted for 73% of all IP traffic, and its share is expected to increase to 82% by 2021 due to the continued growth in streaming video and virtual reality. The main content providers, such as Amazon and Netflix, actively cooperate with companies hosting client servers on their sites to expand their content delivery networks and transfer large-volume video stream sources closer to users in order to reduce costs and delays.

Delivery of traffic to the local data node, which provides storage and processing in close proximity to the source, should be constantly accelerated. In some cases, the local node may be an autonomous data center. More often it will be a 30–300 kW system consisting of racks or rows in an integrated package that can be installed in any environment.

These systems racks in a single package integrate communications, computing and storage with proper power protection, environmental monitoring and physical security. For archetypes requiring high availability, the local node must contain redundant backup power systems and equipment for remote management and monitoring. Many use cases will also require the local host to have data encryption and other security features.