Edge computing is a decentralized information technology (IT) architecture in which client data is processed on the periphery of the network, as close as possible to the original source.
Data is the lifeblood of modern business, providing valuable business insights and supporting real-time control of critical business processes and operations. Today's enterprises are awash in data, regularly collected from sensors and IoT devices that perform on-the-fly from remote locations and inhospitable operating environments almost anywhere in the world.
But this proliferation of virtual data is also changing the way companies handle computing. Traditional computing formalization built in centralized data centers and on the daily Internet is not well suited to the endless growth of mobile data rivers in the real world. Bandwidth limitations, latency issues and unpredictable network outages can all conspire to undermine such efforts. Enterprises are addressing these data challenges by using edge computing architectures.
At its simplest, edge computing moves some storage and computing resources away from central data centers and closer to the source of the data itself. Rather than transmitting raw data to a central data center for processing and analysis, this work happens where the data is actually generated—whether in a retail store, on a factory floor, in a massive utility, or across a smart city. Only the results of this computing work at the edge, such as immediate business insights, equipment maintenance predictions, or other actionable answers, are sent back to the main data center for review and other human interactions.
As a result, edge computing is reshaping IT and business computing. Get a comprehensive understanding of what edge computing is, how it works, cloud implications, edge use cases, trade-offs, and implementation considerations.
Data is the lifeblood of modern business, providing valuable business insights and supporting real-time control of critical business processes and operations. Today's enterprises are awash in data, regularly collected from sensors and IoT devices that perform on-the-fly from remote locations and inhospitable operating environments almost anywhere in the world.
But this proliferation of virtual data is also changing the way companies handle computing. Traditional computing formalization built in centralized data centers and on the daily Internet is not well suited to the endless growth of mobile data rivers in the real world. Bandwidth limitations, latency issues and unpredictable network outages can all conspire to undermine such efforts. Enterprises are addressing these data challenges by using edge computing architectures.
At its simplest, edge computing moves some storage and computing resources away from central data centers and closer to the source of the data itself. Rather than transmitting raw data to a central data center for processing and analysis, this work happens where the data is actually generated—whether in a retail store, on a factory floor, in a massive utility, or across a smart city. Only the results of this computing work at the edge, such as immediate business insights, equipment maintenance predictions, or other actionable answers, are sent back to the main data center for review and other human interactions.
As a result, edge computing is reshaping IT and business computing. Get a comprehensive understanding of what edge computing is, how it works, cloud implications, edge use cases, trade-offs, and implementation considerations.
How does edge budgeting work?
Edge operations are all a matter of position. In traditional enterprise computing , data is generated at the client endpoint, such as the user's computer. This data moves across the WAN through an enterprise area network (such as the Internet), which stores and processes the data in enterprise applications. The results of that work are then passed back to the client endpoint. This remains a proven, time-tested approach to client-server computing for most typical business applications .
But the number of devices connected to the Internet, and the amount of data generated by these devices and used by enterprises, is growing too fast for traditional data center infrastructure to accommodate. Gartner predicts that by 2025, 75% of enterprise-generated data will be established outside of centralized data centers. The prospect of moving so much data, in situations that are often time- or disruption-sensitive, puts incredible pressure on the global Internet, which itself is often subject to traffic jams and interference.
Therefore, IT architects shift their focus from the central data center to the logical edge of the infrastructure—obtaining storage and computing resources from the data center and moving these resources to the point of data generation. The principle is simple: If you can't move the data closer to the data center, move the data center closer to the data. The concept of edge computing is not new, and is rooted in decades-old ideas of remote computing - such as remote offices and branch offices - placing computing resources where they are needed rather than relying on a single central location is more reliable and more efficient.
Therefore, IT architects shift their focus from the central data center to the logical edge of the infrastructure—obtaining storage and computing resources from the data center and moving these resources to the point of data generation. The principle is simple: If you can't move the data closer to the data center, move the data center closer to the data. The concept of edge computing is not new, and is rooted in decades-old ideas of remote computing - such as remote offices and branch offices - placing computing resources where they are needed rather than relying on a single central location is more reliable and more efficient.
Edge computing places storage and servers where the data resides, often requiring only a partial rack to execute on a remote LAN to collect and process data locally. In many cases, computing gear is housed in shielded or hardened housings to protect the gears from extreme temperatures, humidity, and other environmental conditions. Processing typically involves normalizing and analyzing data streams for business intelligence, with only the results of the analysis being sent back to the master data center.
Concepts of business acumen can vary widely. Some examples include retail environments where video surveillance on the showroom floor may be combined with actual sales data to determine the most ideal product configuration or consumer demand. Other examples involve predictive analytics that can guide device maintenance and repairs before actual defects or failures occur. There are other examples that are often aligned with utilities, such as water treatment or power generation, to ensure the installation is performing properly and maintaining output quality.
Edge vs. Cloud vs. Fog Computing
Edge computing is closely related to the concepts of cloud computing and fog computing . While there is some overlap between these concepts, they are not the same thing and generally should not be used interchangeably. It is helpful to compare concepts and understand their differences.
One of the easiest ways to understand the differences between edge, cloud and fog computing is to highlight their common themes: all three concepts are related to distributed computing and focus on the computing and storage resources associated with the data being generated. Physical deployment, the difference lies in the location of these resources.
One of the easiest ways to understand the differences between edge, cloud and fog computing is to highlight their common themes: all three concepts are related to distributed computing and focus on the computing and storage resources associated with the data being generated. Physical deployment, the difference lies in the location of these resources.
Compare edge cloud, cloud computing, and edge computing to determine which model is best for you.
Edge. Edge computing deploys computing and storage resources where data is generated . This ideally co-locates compute and storage with data sources at the edge of the network. For example, a small enclosure with a few servers and some storage might be installed on a wind turbine to collect and process data generated by sensors inside the turbine. As another example, a train station might place a small amount of computing and storage within the station to collect and process data from numerous track and rail traffic sensors. The results of any such processing can then be sent back to another data center for manual review, archiving, and merging with other data results for broader analysis.
Cloud. Cloud computing is a vast, highly scalable deployment of computers and storage resources in one of several distributed global locations (regions). Cloud providers also integrate various pre-packaged services for IoT operations, making the cloud the preferred centralized platform for IoT deployment. But while the cloud provides far more than enough resources and services to handle complex analytics, the nearest regional cloud facility may still be hundreds of miles away from the data collection point, and the connectivity relies on the same temperament that supports traditional data centers. Internet connection. In practice, cloud computing is a replacement for, and sometimes a complement to, traditional data centers. The cloud can bring centralized computing closer to the source of data, but not at the edge of the network.
Cloud. Cloud computing is a vast, highly scalable deployment of computers and storage resources in one of several distributed global locations (regions). Cloud providers also integrate various pre-packaged services for IoT operations, making the cloud the preferred centralized platform for IoT deployment. But while the cloud provides far more than enough resources and services to handle complex analytics, the nearest regional cloud facility may still be hundreds of miles away from the data collection point, and the connectivity relies on the same temperament that supports traditional data centers. Internet connection. In practice, cloud computing is a replacement for, and sometimes a complement to, traditional data centers. The cloud can bring centralized computing closer to the source of data, but not at the edge of the network.
Fog. But computing and storage deployment options are not limited to the cloud or edge. Cloud data centers may be too far away, but edge deployments may simply be too resource-constrained or physically dispersed or dispersed to make strictly edge computing practical. In this case, the concept of fog computing can be helpful. Fog computing typically takes a step back and puts computing and storage resources "into" the data, but not necessarily "in" the data.
Fog computing environments can produce confusing sensor or IoT data over vast physical areas that are too large to define edges. Application cases include smart buildings, smart cities and even smart utility networks. Consider a smart city, where data can be used to track, analyze, and optimize public transportation systems, municipal utilities, city services, and guide long-term urban planning. Single-sided deployment is simply not enough to handle such a load, so fog computing can operate a series of fog node deployments within the environment to collect, process and analyze data.
Note: It is important to repeat that fog computing and edge computing have almost the same definition and architecture, and these terms are sometimes used interchangeably even among technology experts.
Note: It is important to repeat that fog computing and edge computing have almost the same definition and architecture, and these terms are sometimes used interchangeably even among technology experts.