Distributed Cloud: for lower latency and higher cloud service performance. Distributed cloud is a cloud service model that no longer only provides centralized cloud services but brings resources closer to service users. It consists of centralized and decentralized services and ensures optimal utilization of network access options and available transportation bandwidth. Services can be handled in different locations. Distributed Cloud advantages include lower latency, higher performance, and better redundancy. Multi-Access Edge Computing (MEC) and Fog Computing are distributed cloud infrastructure applications. Distributed cloud applications include Internet of Things (IoT), Industry 4.0, Streaming Video, Machine Learning, Artificial Intelligence (AI), Independent Driving, Content Delivery Networks, and others.
Distributed cloud concepts and architecture
One of the main objectives of Distributed Cloud is to bring cloud services closer geographically with users and reduce latency. Distributed clouds are conceptually different in structure. Often, hierarchical cloud concepts are used for realization. Services are provided in this case in various cloud hierarchies. This hierarchy level is:
- central cloud
- some regional clouds
- lots of Edge Clouds
The central cloud provides all services and receives or sends data from regional cloud or edge edge. Cloud regional takes proxy and caching functions and acts as an intermediary between core cloud and cloud edge, or offers their own services. Edge Clouds are placed as close as possible to users and provide cloud services with minimal latency.
Caching as an important technology from a distributed cloud
Many different technologies are used to implement distributed cloud architecture. Caching plays an important role in a distributed cloud. In order not to need to reload data for each request from the central core cloud, but to keep information close to the user, an intelligent caching mechanism is needed. Caching is not just about data caching, but in some cases full service. Caching must ensure that the data and services you need are directly in the edge area. In addition, cached data in hierarchical structures must be constantly synchronized with regional and central cloud services to avoid inconsistencies.
Heterogeneity challenges for distributed cloud
A distributed cloud consists of many different individual components and services. The result is a heterogeneous environment that needs to be mastered. Distributed cloud must build compatibility between services, networks and individual components. Heterogeneity can occur in various areas. These areas are:
- heterogeneous hardware platforms and infrastructure
- network technology, connections and heterogeneous devices
- heterogeneous software implementation
- heterogeneous service providers
Benefits of Distributed Clouds
The Distributed Cloud Concept offers several advantages. Cloud providers use distributed cloud models to bring their cloud services to users with higher performance, better availability, and lower latency. By using a distributed architecture, services can also be improved and adapted to user needs. Good network performance and the service itself can be scaled well. With less data to be sent to the central cloud, the central bandwidth requirements are reduced dramatically. Congestion and congestion can be avoided. This is especially true for applications with high bandwidth requirements such as video streaming.
Regulatory or compliance requirements can also support distributed cloud usage. The Cloud Service can be deployed spatially according to different regional countries or specifications. Service availability increases because the distributed cloud structure continues to function for a period of time regardless of core cloud availability. Applications such as self-driving or automated processes in Industry 4.0 benefit from this, because higher availability ensures greater security or avoids production downtime. Further benefits are:
- Transparent cloud structure for users and works independently of the user's location
- Distributed cloud is ideal for cellular use and offers services that are optimized for each location
- Distributed cloud is suitable for real-time applications because of low latency
- Services can be used in locations determined by the specified Quality of Service
- Examples of applications for distributed cloud
Distributed cloud is suitable for various applications. A typical application is the Content Delivery Network. Decentralization ensures that content, such as high-quality videos, is sent regardless of the user's location. Content delivery solutions function on a variety of network technologies and use distributed storage systems with smart caching technology. They provide high bandwidth for users and reduce the central network capacity needed. Instead of sending and transferring each video stream from the source to the recipient several times, the recording is stored decentralized and sent from there to the recipient.
Examples of other applications are processing and storing personal data.
There are legal or regulatory requirements that prohibit the processing of other personal or sensitive data outside a particular country or region, or only under strict conditions. Distributed cloud can be used to store all affected data in the desired area.
Autonomous driving uses services that process data collected by vehicle sensors and reports the results back to the vehicle to respond to certain driving situations. Distributed cloud ensures that data is processed and sent in real time with minimal latency. In addition, distributed architecture prevents dependence on central cloud components. The situation is similar to Industry 4.0's automatic process data or Internet of Things (IoT). Again, the application relies on processing large volumes of data in the shortest possible time and minimizing service downtime.
