Cloud, edge computing, in the era of the IIoT, at what tier should the data processing be better done ?
Faced with the acceleration of global competition, industries in all sectors are driving the main lever of growth that is emerging: the digital transformation.
While Cloud Computing, the primary vector of this transformation, has become commonplace in all organizations, Edge Computing is emerging as the key enabler to realize the full potential of the Internet of Things.
Cloud and Edge computing are gradually emerging as complementary tiers to provide an optimal level of data processing and storage, according to their purpose and degree of sensitivity.
IIoT or industrial internet of things
Depending on the applications and the flow rate relative to the required data volume, devices communicate via a wireless network (Wi-Fi, mobile phone networks, extended range radio network, long-range network, etc.), which connect them to the Internet or to a local network.
At the same time, talking about Industry 4.0, the industry of the future, or smart manufacturing, … is a way of talking about the digital transformation of the global operations of industrial companies in the era of Industrial Internet of Things (IIoT).
This transformation, considered the fourth industrial revolution, is a mandatory step to support competitiveness and corporate leadership. It also offers a tremendous potential for innovation, profitability and growth.
It is for manufacturers to use networks of intelligent objects to increase the profitability of activities by improving the entire value chain through asset monitoring, quality control, productivity optimization, or performance of predictive maintenance.
According to PWC’s “Industry 4.0_ Opportunities and Challenges of the Industrial Internet” study, by 2020, 1 billion connected objects will have invested the factories, 80% of companies will have digitalized their value chain, and productivity will grow by 18%.
Cloud computing for more agility
The data is centralized in delocalized data centers, often redundant for security reasons, which offer a computing and storage capacity superior to the capacities conceivable on industrial premises.
Through cloud computing, industries have freed themselves of the cost and the burden of hosting their servers and applications internally while having access to their exploitable data anywhere, anytime and at a lower cost.
Particularly suitable for the centralized and secure management of huge volumes of information, its use also induces a virtual and physical round-trip time which is more or less sensitive depending on the criticality of the concerned data.
At the same time, billions of connected objects to which we are aiming will generate a colossal amount of data that cannot and should not be handled by traditional networks in terms of both data flow and response time. It is clear that the issue of latency in data recovery and exchange is a crucial issue.
Edge computing for more responsiveness and efficiency
According to wikipedia, in computing, a gateway is the generic name of a device for connecting two computer networks of different types, for example a local network and the Internet.
Data is the cornerstone of informed decision-making, and Edge Computing is precisely the data processing tier that allows the quickest access to information as it is done in the sensor or gateway at the edge; closest to the data collection point. This fast-on-site processing eliminates roundtrips with the cloud.
The computing power is deported at the level of the sensor network, to extract the value of the data in real time. The information needed for the action / reaction is immediately available which increases the power of the IoT for applications where the real time is the crux of the matter.
It is in essence the ideal model for processing the most critical data in terms of time. These data are captured and pre-processed or processed on board, on the sensors themselves, “at the edge of the network”, leaving the posterior analyzes to the cloud computing level.
The computing power found in Edge computing offers greater efficiency for the provision of critical information, immediately useful. Gateway processing and storage reduces the amount of data in transit to the cloud and reduces bandwidth. The network is decongested, more reliable, and only information that contributes to in-depth analysis is transferred to a centralized data center.
Edge computing delivers the efficiency and responsiveness needed for real-time critical data. By placing computing resources at the edge of the network, it guarantees the operation of differentiated sites, independently of possible failures at the base, and offers both reliability and resilience.
Correlatively, it meets the requirement of less pollution required to offset the exponential growth of information flows from the Internet of Everything.
According to Trendforce, the annual growth rate of edge computing will exceed 30% from 2018 to 2022.
Cloud and edge computing the winning alliance
In an ideal world, data processing is prioritized so that the time required to obtain crucial information is minimized and traffic on the network infrastructure is organized to be as light as possible. This organization is based on intelligent objects adjusted to the application needs, a suitable communication network and a relevant platform offering analytics and data visualization tools to fully benefit from the power of the IIoT.
This processing infrastructure, reduces the cost of data hosting and storage, offers the appropriate processing speed depending on the required data types and their purposes, ultimately resulting in optimized energy management.