What is Edge Computing?

What is edge computing?

Edge computing is a computing approach in which collected data is processed at the periphery, or edge, of a network rather than being sent to a centralized server for processing and storage. The point of edge computing is to process data at its originating source, or as close as possible. 

For instance, a surveillance camera with edge computing capabilities can capture data and process it immediately onsite instead of simply relaying it to a server at headquarters for later processing and analysis. 

Organizations are increasingly using edge computing to speed up decision making and response times to incidents. It enables businesses to benefit from real-time insights and alleviates the computing burden on centralized data centers. Processing data at the edge means the device doesn’t send great amounts of raw data to central servers—it sends only the insights and analysis for further use.

How it works

Before edge computing, all data produced at an endpoint—be it an employee’s office workstation or a surveillance camera in a remote warehouse—had to be relayed back to a data center to be processed and stored. Applications would analyze the data, gather insights, and then (when appropriate) send information back to the device. 

Edge computing keeps data processing fast and efficient: 

1. A device gathers data in a location outside an organization’s office, whether it’s a surveillance camera in a remote warehouse or a location sensor on a delivery truck. 
2. Instead of relaying raw data over the network to a central data center for processing, the device has built-in computing capabilities. The data is processed as it is collected, so anomalies or other issues are identified in near real-time. 
3. The results of the analysis can then trigger actions from the device, such as sending an alert or sounding an alarm. 
4. Only the results and insights are sent back to the central data center. This reduces network traffic and latency. 

See how Nutanix Cloud Platform can support your edge computing objectives by forming a smart foundation for your cloud infrastructure.

Cloud computing vs fog computing vs edge computing

Because edge computing relies on distributed IT architecture, it is sometimes used interchangeably or confused with cloud computing or fog computing. While these three approaches have some similarities, such as their distributed architecture and the placement of storage and compute resources nearer to the data’s origination point, they are not the same thing.

Cloud computing is enabled by a massive collection of servers located around the world. When you use AWS, for instance, your data is stored and processed in one of their data centers instead of your own on-premises infrastructure. It could potentially be closer to the data origination point than your data center is, but it’s not at the edge. Collected data still has to be relayed to one of these data centers for analysis.

Fog computing is helpful in situations where edge devices are located across a very large area, such as in a smart building. There could be hundreds of edge devices in that environment and the data they collect needs to be aggregated, processed, and analyzed together to get the best results. So fog computing allows you to place storage and compute resources within that environment but separate from any single device—because no device could contain enough computing power to handle data from all the others.

Benefits of edge computing

• Increased operational efficiency – By accelerating time to insights and processing data in near real-time at the edge, you get results faster. It also reduces network latency and eases bandwidth concerns. 
• Faster issue resolution – With insights from the edge, you can receive alerts about potential issues before they escalate and cause excessive downtime or other problems. 
• Lower IT costs – Edge computing allows you to reduce the necessary compute and storage resources in your data center, which helps save on costs. Reducing the amount of data transferred from edge devices also lowers network costs. 
• Increased data security – Data gathered and processed at the edge can help you adhere to data privacy laws, where moving it to a central data center could affect data privacy and sovereignty. Keeping data on the edge device also helps protect data from interception over the network. 
• Increased reliability – Edge devices can continue to collect and process data even when your on-premises data center or network is down. This helps enhance business continuity.

Challenges of edge computing

While edge computing accelerates data processing and decision making, it does come with some challenges. These include:

• Increased infrastructure complexity – An organization’s edge devices can number in the thousands and beyond, and with each device comes more need for maintenance and management. That means an increased burden of software updates, deployments, provisioning, and monitoring.

• Security concerns – While edge computing can increase data security in some ways, device security is a bit more complicated with large numbers of devices.

• Connectivity issues – Even though they’re not transmitting raw data to a centralized location, edge devices still do have to relay analyzed data back to some servers. If the internet goes down where the device is, that can spell trouble for an organization that relies on timely delivery of those insights.

• Storage efficiency – Because edge devices have limited compute and storage resources, sometimes IT will have to determine which data to store and process locally and which data to send to on-premises servers or the cloud.

Fortunately, there are a wide range of solutions available today that are designed to help organizations overcome these edge computing challenges.  

Nutanix has advanced solutions that can make your edge deployments simpler to manage and control. With a commitment to enabling organizations to have data and applications right where they need them—whether on-premises, in the cloud, or at the edge—Nutanix offers a a hybrid multicloud platform that makes it easy to monitor and manage your workloads anywhere and everywhere. With the Nutanix Cloud Platform, you get a fully unified foundation that connects every environment together for seamless mobility and accessibility.

Edge computing use cases

As edge computing gains popularity, more use cases emerge. They include:

• Autonomous vehicles – Self-driving vehicles can generate multiple terabytes of data every day. That data needs to be processed and analyzed in real time to make split-second safety decisions. In this case, each vehicle is an edge device and requires robust compute and storage resources.
• Manufacturing – From safety on the production floor to identifying sub-par products on the assembly line, edge computing has quickly taken hold in manufacturing environments.
• Oil & gas – Remote oil rigs in the ocean rely on edge computing to alert workers of equipment malfunctions before they become serious enough to threaten lives.
• Healthcare – Many healthcare organizations are turning to edge computing to track and maintain hospital equipment such as wheelchairs, alert doctors to potential health issues through patient equipment such as CPAP machines and vital sign trackers, and increasing safety in hospital locations via surveillance devices and alerting systems.
• Retail – Retailers benefit from edge computing by using it to track inventory, surveil storage warehouses, analyze customer behaviors, and more.
• Energy – Edge computing can help energy companies manage power grid automation, monitor the condition of assets in very remote locations, and alert teams to potential equipment malfunctions.