A report on Edge Computing from Gartner, may give you a deeper understanding of Edge Computing and inspire you how to respond to the development of Edge Computing.
Get the complete report here: >> “Hype Cycle for Edge Computing”
Edge Computing is regarded by many as an exaggerated, fuzzy-defined computing model, usually related to the Internet of Things. In fact, Edge Computing concepts are extracted from proven distributed computing models (such as client servers), which are described as a two-way evolution of content delivery networks. Edge Computing is defined as “part of a distributed computing topology where information processing is located near the edge-where things and people produce or consume information.
Although Edge Computing has just started, as a decentralized supplement to the rapidly mature super-large-scale cloud service, and the symbiotic relationship between the edge and the Internet of Things, its development speed will be very fast. For example, IoT vendors have been building endpoint solutions for many years. Hyperscale cloud providers, especially AWS, Microsoft and Google, have also invested a lot of resources to actively bring edge support software to the market.
Technologies or applications that are expected to mature in the next 2 to 5 years include:
- Edge AI
- Edge Computing
- Edge Video Analytics
- Micro Data Centers
Especially worthy of our attention is the Micro Data Centers, as a key basic support for Edge Computing, Gartner predicts that it will mature in the next 2-5 years and can be applied on a large scale.
Definition of Micro Data Center:
A Micro Data Center is modular or containerized and is smaller than a computer room-usually no more than one equipment rack or two-usually one rack or less. All necessary facilities and IT functions (such as uninterruptible power supply, servers, storage, networking, and cooling) are included in the Micro Data Center. It is designed to handle the specific needs of distributed locations (for example, accumulating sensor data or small remote office support), and is usually managed remotely from large data centers.
Location and Reason:
For more than 20 years, small computer rooms (small as closets) have been merged into larger, integrated data centers, or their workloads have migrated to external cloud providers, and we see this trend still Continue. However, not all IT functions need (or should) be centralized. Placing workloads locally on business units or on factors such as latency or site-specific operations are common reasons for deploying Micro Data Centers. In addition, due to the distributed nature of the Internet of Things (IoT), it may be necessary to accumulate and process data locally before sending a subset of data to a larger data center for additional analysis. Micro Data Centers are usually based on more mature IT and facility technologies, although new features have been added, including hyper-aggregation, packaging, containerization, and remote management. Trends such as the IoT and the distributed nature of Edge Computing mean that the deployment of Micro Data Centers will be much broader than it is today. They can be distributed almost anywhere, including as an independent center, placed in a suitable container (for environmental protection), usually found in branches. Micro Data Centers use any available means of communication, but are usually connected to the office LAN / WAN.
With the rich R&D and practical experience in Micro Data Centers, Attom Technology is continuously supporting the better and faster development of Edge Computing with innovative Micro Data Center products.
Are you interested in Micro Data Centers? Please contact us.