{"id":38375,"date":"2019-10-31T22:23:21","date_gmt":"2019-10-31T19:23:21","guid":{"rendered":"https:\/\/prohoster.info\/blog\/iot-tuman-i-oblaka-pogovorim-pro-tehnologii\/"},"modified":"2019-10-31T22:23:21","modified_gmt":"2019-10-31T19:23:21","slug":"iot-tuman-i-oblaka-pogovorim-pro-tehnologii","status":"publish","type":"post","link":"https:\/\/prohoster.info\/en\/blog\/news\/iot-tuman-i-oblaka-pogovorim-pro-tehnologii","title":{"rendered":"IoT, fog and cloud: shall we talk about technologies?","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

\"IoT,
\n
\nThe advancement of technology in software and hardware, along with the emergence of new communication protocols, has led to the expansion of the Internet of Things (IoT). The number of devices is growing day by day, generating vast amounts of data. This creates a need for a convenient system architecture capable of processing, storing, and transmitting this data.<\/p>\n

Currently, cloud services are used for these purposes. However, the increasingly popular paradigm of fog computing can complement cloud solutions by scaling and optimizing the IoT infrastructure. <\/i><\/a><\/noindex><\/p>\n

Clouds can meet most IoT demands. For instance, they can provide monitoring services, rapid processing of any data volumes generated by devices, as well as their visualization. Fog computing, on the other hand, is more effective in solving real-time tasks. It ensures quick responses to requests and minimal latency in data processing. In other words, fog computing complements clouds and expands their capabilities.<\/p>\n

However, the main question is: how should all of this interact in the context of IoT? Which communication protocols will be most effective in operating within the unified IoT-Fog-Cloud system?<\/p>\n

Despite the apparent dominance of HTTP, many other solutions are used in IoT, Fog, and Cloud systems. This is due to the fact that IoT must combine the functional capabilities of various device sensors with security, compatibility, and other user requirements.<\/p>\n

There is simply no uniform understanding of a reference architecture and communication standard. Therefore, creating a new protocol or adapting an existing one for specific IoT tasks is one of the most important challenges facing the IT community.<\/p>\n

What protocols are currently in use and what can they offer? Let's find out. But first, let's discuss the principles of the ecosystem in which clouds, fog, and the Internet of Things interact.<\/p>\n

IoT Fog-to-Cloud (F2C) Architecture<\/h3>\n

\nYou have likely noticed the significant efforts put into exploring the advantages and benefits associated with the rational and coordinated management of IoT, clouds, and fog. If not, here are three standardization initiatives for you: OpenFog Consortium<\/a><\/noindex>, Edge Computing Consortium<\/a><\/noindex> and mF2C H2020 EU project<\/a><\/noindex>. <\/p>\n

While earlier discussions focused solely on two levels, the cloud and end devices, the proposed architecture introduces a new tier\u2014fog computing. This fog level can be subdivided into several sublevels, depending on the specific resources or sets of policies that define the use of different devices within these sublevels.<\/p>\n

What could this abstraction look like? Here is a typical IoT-Fog-Cloud ecosystem. IoT devices send data to more powerful servers and computing devices to solve tasks that require low latency. In this system, clouds are responsible for handling tasks that require a large amount of computational resources or data storage space.<\/p>\n

\"IoT,
\n
\nSmartphones, smartwatches, and other gadgets can also be part of IoT. However, such devices typically use proprietary communication protocols from major developers. The data generated by the Internet of Things is transmitted to the fog level via the REST HTTP protocol, which ensures flexibility and functional compatibility when creating RESTful services. This is crucial in light of the need to maintain backward compatibility with existing computing infrastructure that operates on local computers, servers, or server clusters. Local resources, called 'fog nodes,' filter the received data and process it locally or forward it to the cloud for further computation.<\/p>\n

Clouds support various communication protocols, with AMQP and REST HTTP being the most commonly encountered. Given that HTTP is well-known and designed for the Internet, one might wonder, 'Should it be used for working with IoT and fog?' However, this protocol has performance issues. More on this later.<\/p>\n

In general, there are two communication protocol models suitable for our system. These are request-response and publish-subscribe. The first model is more widely known, especially in client-server architecture. The client requests information from the server, which receives the request, processes it, and returns a response message. REST HTTP and CoAP protocols operate according to this model.<\/p>\n

The second model emerged from the need to provide asynchronous, distributed, loosely coupled connections between data-generating sources and their recipients.<\/p>\n

\"IoT,
\n
\nThe model involves three participants: the publisher (data source), the broker (dispatcher), and the subscriber (recipient). Here, the client acting as a subscriber does not need to request information from the server. Instead of sending requests, it subscribes to certain events in the system via the broker, which is responsible for filtering all incoming messages and routing them between publishers and subscribers. When an event related to a particular topic occurs, the publisher publishes it to the broker, which sends the data to the subscriber on the requested topic.<\/p>\n

Essentially, this architecture is event-driven. Such an interaction model is particularly interesting for applications in IoT, cloud, and fog computing due to its ability to ensure scalability and simplify relationships between various devices, enabling dynamic 'many-to-many' and asynchronous communication. Some of the most well-known standardized messaging protocols that utilize the 'publish-subscribe' model include MQTT, AMQP, and DDS.<\/p>\n

It is clear that the 'publish-subscribe' model has numerous advantages:<\/p>\n