Taking computing to the edge

The intelligent edge is defined as the place where data is analysed and aggregated in a spot close to where it is captured in a network, in a lot of cases, on a mobile device. This is the next step in the evolution of cloud computing.

Instead of decision-making happening in large-scale data centres, as it does today, they happen on devices which sit at the edge of the cloud. This creates efficiencies, speeds up reaction times and solves customer issues in real-time. Because data isn't shipped off to a data centre or third party for analysis, it increases the speed of said analysis. Also, data is less likely to be intercepted, reducing the security risks to proprietary data. Key advantages of the intelligent edge are the reduction of latency, data processing costs, and security risks.

There is still much debate as to when and how the intelligent edge will become the norm in South Africa, but there are a few things that will determine this, says Basha Pillay, Executive Head: Cloud and Collaboration at Internet Solutions.

They are:

1. The ability of Internet service providers to provide the requisite infrastructure and support necessary for edge computing;
2. Consumer demand;
3. The proliferation of connected devices and the Internet of things; and
4. The development of chip technology that facilitates the processing of data at the intelligent edge.

Role of ISPs at the edge

The intelligent edge will mean less load on the network, and we will see an increased demand for a distributed data warehouse model. We will see core intelligence located centrally and regionalised data centres housing improved data sets in closer physical proximity to users. Data required for quick decision-making will be processed at the edge, for example, data from IOT sensors in use for smart city management. Networks will be characterised by a hybrid system of micro and macro data centres.

ISPs will continue to play an integral role in connecting devices and providing the platform on which computing at the edge can occur.

Of major importance will be providing the requisite security to ensure data remains safe. ISPs will be responsible for data storage and processing, including protecting that data at the edge as well as the cloud itself. Capacity for machine learning, to facilitate edge decision-making, will also be the responsibility of ISPs.

Consumer demand

It's estimated that by 2025, 60% of cloud services globally will be located at the edge to meet consumer expectations for low latency, high-demand applications like content consumption. Africa has already demonstrated an appetite for new technologies that solve long-term issues, especially in the financial services industry.

A study conducted by Avaya in 2017, The Customer Experience in Banking Survey, revealed that South African consumers prefer a digital-first approach to resolving their basic banking issues. The survey queried more than 10 000 banking customers in nine countries. Of the nine countries surveyed, South Africans were the most likely to say they preferred to use a mobile banking app to access their banking services.

In this mobile-first environment, the natural next step is the adoption of intelligent edge solutions to meet consumer demands for instant service and satisfaction.

Connected devices

Parallel to the role of consumer demand driving the need for intelligent edge is the spread of connected devices as part of the proliferation of the Internet of things. The IDC says by 2025, approximately 80 billion devices will be connected to the Internet globally. Each one of those devices will have the ability to speak to each other and collect massive amounts of data about users. This data will be stored centrally; however, the edge will mean less reliance on these centres to make key decisions.

This lack of reliance on major data centres for processing power will mean a move away from centralised computing and macro data centres to micro or regional data centres. However, it is likely the decisions being made at the edge will initially be mundane and still require the artificial intelligence capabilities of hyper-scale data centres.

Silicon at the edge

Chip manufacturers will be another major driver of the pace of edge computing, not just locally but around the world. The greatest impact of this move will be felt in the development space, where we will begin to see decisions being made at a sensor level.

In February, Intel unveiled a new chip designed specifically for computing at the edge, the Intel Xeon D-2100 processor. Other chip producers have followed suit, announcing AI and IOT-enabled chipsets as well. These new technologies will play a vital role in bringing the cloud to the edge and making smart devices even smarter.

The future of cloud technology is at the edge. As the IOT and connected devices become more commonplace, consumers are beginning to demand quicker solutions to their problems. In an interconnected world this means cutting down on processing times to react faster to changing environments and demands. When edge technology arrives in South Africa, it will bring with it huge opportunities to improve people's lives and drive efficiencies in business. For ISPs, it will present an opportunity to provide their services in new markets and across new computing models.