When Intelligence moves to Edge – The proliferation of the Internet of Things (IoT) and the rich growth of the cloud services have pushed the horizon of a new computing paradigm which processes the data at the ‘Edge of the Network.’
Edge computing may sound like some hyper-advanced technology, but it is not. It refers to the location where data is processed. Think of a data centre as the hub of a computer network and outside offices or business locations as network edges. Edge computing is the practice of collecting and analyzing data where it is generated – At network edges. Data generated at network edges can come from a variety of sources – Phones, IoT devices, POS systems – anything that records and transmits data. As more internet-connected devices become part of the business, they generate more data. Sending all that raw data from the Edge to the hub becomes more difficult.
Edge computing shifts analysis to on-site machines and only transmits results, and lots of businesses are already doing it. Corporations use it to improve security. Semiconductor manufacturers use it to monitor chip quality. Mining companies even use it to track truck locations to prevent black market resale. The quantity of data in the world is only growing to grow, and so will the practicality of edge computing.
The evolution of the Internet of Things (IoT) and advances in sensor technology, data is being collected further and more from the confines of a traditional data centre. That data needs to be processed increasingly at, or near, the location where the data is being collected. This is where the edge computing comes into existence by bringing computing power to the data-collection devices. The systems can take advantage of lower latency to provide near real-time insights for users. This could help speed up data authentication, analysis, and more. However, the Edge can also be further used to filter data sets so that only the relevant data is sent to the cloud or data centre for additional processing.
Locating the Edge:
Today, there is a pent-up data in things, such as even cat or dog food bowls can tweet you when they are empty. It is data, but how valuable is it? Guess it is essential for the cat/dog. Edge computing model is vitally necessary for capturing, processing, and managing the enormous amount of static data, which is inevitable to be produced. “An edge is a place,” It is where the things are, and it is not the data centre or the cloud. The Edge will start housing the right IT products that are in data centres and clouds.” One class of these products is what HPE calls operational technology (OT), which one might interpret to be a large slice of the market that incorporates data centre infrastructure management (DCIM) and systems monitoring.
Different types of edges:
There are three broad categories of Intelligent Edge: Operational technology (OT) edges, IoT edges, and Information technology (IT) edges.
- Operational technology edges: OT edges commonly contain intelligence and controls but have been traditionally limited in connectivity and compute power. OT edge examples include offshore oil rigs and power plants.
- Information technology (IT) edges: IT edges are typical in the Media and Telecommunications industries for distributed data transfer and processing, as well as distributed computing in branch offices and campuses.
- IoT edges: The IoT edge is of a great interest today, as the IoT is making everything connected in our world. In many instances, the IoT is a combination of OT and IT.
When discussing the IoT, it’s instructive first to understand the generalized four-stage; IoT solutions architecture depicted below. “Things” are connected to sensors for data capture and actuators to control the things—either wired or wirelessly. These sensors and actuators connect to gateways, switches, and data acquisition systems in stage 2. Stage 3 is comprised of IT systems that are at the Edge, and stage 4 is the remote data centre or cloud. Not all IoT solutions include all four stages (e.g., some are sensor-to-cloud solutions), but a large portion of IoT solutions can be mapped into this architecture.
Industrial IoT (IIoT) solutions can generate significant business value when sensors are connected to rotating machinery such as turbines in an electric power plant. Physical data extracted from the turbines, such as temperature, vibration and moisture provides valuable insights into the machines’ health. Prognostic analytic programs process this data and provide immediate insight into the turbines’ health status. This affords better control over the maintenance routines and helps predict failure associated with burnouts or blackouts.
Edges around Us – Edge Computing examples can be increasingly found around us in:
- Smart street lights
- Automated Industrial Machines
- Mobile devices
- Smart Homes
- Automated Vehicles (cars, drones etc.)
Data Transmission is expensive. By bringing compute closer to the origin of data, latency is reduced as well as end users have a better experience. Some of the evolving use cases of Edge Computing are Augmented Reality(AR) or Virtual Reality(VR) and the Internet of things (IoT). For example, the rush with which people got while playing an AR-based pokemon game, wouldn’t have been possible if “real-timeliness” was not present in the game. It was made possible because the smartphone itself was doing AR, not the central servers.
Even Machine Learning(ML) can benefit greatly from Edge Computing. All the heavy-duty training of ML algorithms can be done on the cloud, and the trained model can be deployed on the Edge for near real-time or even real-time predictions. We can see that in today’s data-driven world, edge computing is becoming a necessary component by itself. However, there is a lot of confusion between Edge Computing and IOT. If stated, in a way, Edge Computing is nothing but the intelligent Internet of things(IoT). Edge Computing compliments traditional IoT. In the conventional model of IoT, all the devices, like sensors, mobiles, laptops etc. are connected to a central server. Now let’s imagine a case where you give the command to your lamp to switch off, for such simple task, data needs to be transmitted to the cloud, analyzed there and then the light will receive a command to switch off. Edge Computing brings computing closer to your home, that is either the fog layer present between lamp or cloud servers is smart enough to process the data or the lamp itself.
The significant advantages of computing at the Edge are:
Although all three Cs (connect, compute, and control) contribute to edge intelligence, compute improvements are especially important because they can yield immediate insights from edge data at relatively low cost. Edge computing can be improved by shifting enterprise-class compute, storage, and management from the data centre to the Edge. Organizations can leverage compute at the Edge to:
- Minimize latency: Many applications require immediate insight and control. For some mission-critical functions, computing must happen at the edge level because any latency is intolerable. Consider a robot arm in a shop process that requires precision adjustments and calibration to maintain product quality. If the factory is churning out 50 products per minute, it is critical the calibrations be done in real time to minimize defects.
- Reduce bandwidth: Sending big data back and forth from things to the cloud can consume enormous bandwidth. Edge computing is the most natural solution to this problem.
- Low cost: Even if bandwidth is available, it can be costly. Efficiency is a crucial element of any corporate IoT strategy.
- Reduce threats: When the data is transferred across the state, country, or ocean, it is merely more prone to attacks and breaches. Processing data at the Edge can reduce security vulnerabilities.
- Avoid duplication: If all the data is collected and transmitted to the cloud, there will likely be some equipment duplication in memory, storage, networking equipment, and software. If this duplication is not needed, then the associated increase in capital and operating expenditures are unwarranted.
- Improve reliability: Even without any nefarious activity from hackers, data can be corrupted on its own. Retries drop, and missed connections will plague edge-to-data-centre communications. Even today, cell phone calls can be broken up or dropped.
- Maintain compliance: Laws and corporate policies govern the remote transfer of data. E.g., certain countries forbid companies from moving the personal data of their citizens outside their borders.
To summarise, Edge is the future of computing technology. Edge is the only way for addressing the imminent ‘Data Deluge’ which offers a distributed computing power that moves data centre resources to the network periphery.
Image Credit: zdnet