Always Near the Data Source – Edge Computing and Cloud Services

Author: Aleksandra Dabrowska, Comarch SA

The need to instant access to data has become more crucial for any organization. In manufacturing, retail, telecommunications and healthcare, business processes are becoming ever more automated. This means increased demand for computing powers, cloud solutions and infrastructure that is capable of processing massive quantities of data. More data means better understanding of industries and clients but, at the same time, it can lead to network overload and latency. Cloud computing has already become a standard for enterprises. However, it is not the single best solution when the amount of data flowing to the data center is huge, or when processing has to be nearly real time. The solution is cloud edge computing.


What is Cloud Edge Computing?

Edge computing is an evolution of cloud computing rather than a new solution. It is a network of edge data centers that processes the critical data at the edge of a network, close to the source of data, rather than in the central data repository. The greater part of the network’s computing powers is moved to the edge (capable of performing more complex operations than the data-producing device itself), which decreases latency, speeds up communication and frees up main data repository resources (only part of the pre-aggregated data is sent to the central storage center). In this, way organizations are able to analyze the necessary information in almost real time, which is absolutely crucial in, for example, healthcare, where seconds may decide between life and death. We will present more examples later in the article.

So how does edge computing work? A device (such as a smartphone, laptop, POS, IoT beacon or medical equipment) processes some of the data. If more computing power for processing or archiving is required, data are sent to the edge/cloudlet. The edge data center processes some of the data and, if it is sufficient, those data are not sent to the central cloud unless it is for analytical/archival purposes, in which case the central cloud receives data in a highly aggregated form. This is illustrated below.

Edge Computing’s influence on cloud-based services

Nowadays, around 10% of enterprise-generated data are created and processed outside a traditional central data center or cloud. Gartner predicts that, by 2022, this figure will reach 50%.[1] This allows companies with massive amounts of data-generating machines nearly real-time data processing. It is mostly used in the manufacturing, retail, healthcare, finance, banking and telecommunication industries, where time matters and there is no room for latency or overloaded networks.

Processing data in locations such as shopping malls (with no physical space to keep the IT infrastructure), research centers in deserts, forests or on remote islands would be time-consuming, very expensive and sometimes even impossible if the information had to be sent from the devices to the main cloud thousands miles away. So, it is more practical to do it at the edge of the network and transfer the validated data to the cloud.

In the telecommunications industry, next-generation 5G cellular networks are an important use case. The providers will have to build or rent mini data centers near 5G towers to ensure the stability of cellular and Internet network connections for their customers.

The IoT industry has grown so fast over the past few years that it is slowly reaching network physical limits. Big factories, manufacturers and producers automate their processes by using smart machines filled with IoT technologies. They need to provide and analyze data in real time, which overloads bandwidth and causes delays and errors. By reducing the distance and moving data processing closer to the user, latency can be avoided.

The list of areas that benefit from edge computing is quite long. We will give just a few examples:

  • Logistics (tracking and monitoring shipments in real time)
  • Retail (beacons communicating with clients via location and preferences)
  • Manufacturing (switching off malfunctioning machines, controlling production)
  • Health monitors and wearable life bands
  • Security cameras and monitoring systems (facial recognition systems)
  • Smart traffic lights (controlling the balance between vehicles and pedestrians)
  • Smart home equipment (smart air conditioning/heating, vacuum cleaners, baby monitors, home assistants, etc.)


Being closer to the source means more benefits

The main, and probably the most important benefit of edge computing is reduction of latency – data do not have to travel to a data center or cloud for processing and back again. Therefore, edge significantly speeds up processing, speeding up decision-making and action, increasing the overall user experience, and lowering bandwidth requirements for Internet connections. That is why the edge computing model works well in places where access to the Internet is either difficult or unreliable. Moving data processing closer to where the data are actually created reduces the risk of downtime caused by network connection failures.

Since data are not being transferred million miles away, the security of information is increased and the risks of security being breached or data compromised are reduced to minimum. This is crucial in the invoice exchange clearance model (used in Russia and China), where the tax administration office requires each invoice to be reported and authorized electronically by officials during the business exchange process.

Another advantage is the reduction of computing power requirements in the central system. It is possible to use edge data centers precisely where they are needed around the world. This affects the overall cost of equipment necessary to run the whole operation, including maintenance fees and energy consumption.


Comarch is always near the client

Comarch is expanding its network of owned and partner data centers around the world, so that they are more decentralized and closer to clients and their devices, regardless of their location. The local data centers are cheaper to establish, ensure additional security and privacy measures, and are less prone to attacks. Because less data is sent to centralized cloud platforms for analysis, network traffic and data transmission delays are reduced – everything the client could wish for. You can find Comarch Data Centers all around the globe. See where our locations are

Picture 1 Comarch Data Center Maps

Our IT specialists and consultants already have a wealth of experience in implementing edge computing solutions. Comarch has operated successfully in Russia (taking into account that country’s clearance invoice exchange model and the legal regulations that come with it), France, Germany, the United States, Canada and a few Asian countries (where local data centers were created to overcome long distances and sensitivity to delays).

Comarch has also worked with many different industries. Retail projects need edge centers in a few geographical locations, in order to be closer to their customers, because shops are practically everywhere. For Auchan, hosting for European customers is in Poland. The X5 Retail Group, owner of supermarket chains, has its loyalty management handled by an edge center in Russia to ensure flawless operation of services.

Telecommunications companies face with the same issue, but on a bigger scale. To provide customers with a satisfactory signal, edge centers must be located as close as possible.

Comarch helped Nicols, the Belgian producer of cleaning products and accessories, to relocate factories to Poland despite their end users remaining in Belgium. How? Although the manufacturing process was transferred, edge computing meant the IT infrastructure could stay in Belgium, close to Nicols’ clients. In this way, the user experience remained excellent, and there was no clash between production, IT infrastructure and client service.

?ukasz ?acniak, Business Solutions Architect, reveals Comarch’s plans for the foreseeable future:  The first Comarch Enterprise Cloud region has just been launched in our Polish data center. The next step is to finish work there and launch for another region, in the French data center, by the end of 2018. Comarch will not stop there. We intend to expand our portfolio of data center services and introduce hosted private cloud services in Germany and the USA. This is strategically important for Comarch, because the companies we cooperate with are mostly global. The business systems of our clients’ employees and customers are usually transactional systems, critical for the company’s operations and sensitive to delays. In addition to the technical aspects of locating Comarch Enterprise Cloud platform closer to our clients, we also address legal issues – our clients’ data are always stored and processed in accordance with local legal regulations and clients’ preferences and needs.

Varied and flexible cloud solutions are currently available for business. Cloud edge computing can be the technology that changes the way we think about data processing. It offers maximum speed, reduced costs and data security. Of even greater importance, it balances out the computing capabilities of already exploited data centers and clouds. With the development of IoT technology, the 5G network and smart cities, edge computing will be one of the biggest IT trends for years to come.


You must be logged in to post a comment Login

Leave a Reply

To Top