The peripheral environment actually refers to the problem of data management, i.e., the solution of the questions: what data should be stored and which should be discarded after use, which should be analyzed, which should be transferred, and how to ensure security.

To make decisions, enterprises must correlate the cost of peripheral computing with the cost of data transfer. Cost can be expressed in various ways: this is not only an obvious monetary cost, but also time - in fact, when performing time-critical operations, transferring large amounts of data for analysis seems unrealistic, in this case, localization of computing capabilities is required. This implies the need to optimize server selection depending on the specific area of use of peripheral environments.

Retail

Intelligent retailers use different tactical approaches: from mass marketing to personalized service and from mass advertising to multi-channel consumers. A multi-channel customer is one who uses various points of contact before making a purchase — through the Internet, voice assistant, digital channels, email, and in-store purchases.

This is how the requirements for computing, storing and transmitting data are formed for the peripheral retail environment in order to provide differentiated customer service. Trading floors, despite their convenience for random viewing and optimizing the movement of goods, can be fraught with problems in terms of computing environment. A server whose design depends on stable power consumption, cooling and sufficient space for maintenance in a traditional data center will not be the best choice in a retail environment.

Motor transport

It is no secret that the automotive industry is undergoing profound transformations under the influence of IT. These changes include additional computing, networking and storage capabilities not only in the vehicles themselves.

Realizing the capabilities of an autonomous vehicle at a hierarchically higher level in relation to an already connected ecosystem will significantly increase the amount of data coming from radar sensors, GPS, lidars, ultrasound and video subsystems. The inclusion of infotainment options and a list of the right types of computing, storing and transmitting data in all the right places across the peripheral environment is becoming vital to raising awareness, coordination, synchronization and movement of data.

The peripheral environment of vehicles combines several requirements and considerations described in examples of the use of peripheral environments in retail and industrial production. Servers must have I / O flexibility, multiple processors, provide distributed control, work with limited power consumption and in potentially difficult conditions, and be integrated into wider solutions.

Search: selection

Edge computing server selection algorithm