Edge computing
refers to the enabling technologies allowing computation to be performed at the edge of the network, on downstream data on behalf of cloud services, and upstream data on behalf of IoT services. No matter how large or small the computing unit is, mitigating computing tasks from the cloud to any computing resources close to where data is generated accelerates the data analysis and enables the action on insights faster.
What are the Benefits of Edge Computing?
With the coming tsunami of IoT data, solely relying on a centralized computation on cloud and today’s Internet network is not going to cut it for some time-sensitive applications. Edge computing is a viable solution to these challenges as its concept of locating data processing closer to its originating source can speed up response time and reduce data transmission costs. The edge computation result can be fed back to the downstream devices/system for real-time control or sent to the cloud for further review or dissemination. Apart from faster response time, there are additional benefits to edge computing that might interest you to explore this technology, i.e.,
–Data privacy and sovereignty. Offloading data to process or store on the cloud usually means transmitting data across national boundaries which can cause concerning security/privacy/legal issues.Data stored in the cloud is subject to the laws and regulations of the countries in which the data centers reside. Managing data stored in different geographic locations can be confusing. In addition, without end-to-end encryption scheme your data can be exposed to the cloud provider. Therefore, processing sensitive data within trusted infrastructure such as local edge computing devices is not only advisable but critical.
–Offline availability. If every computation is delegated to the cloud, the connectivity to the cloud must be 100% available and reliable. When this is not such a case such as in remote environments, edge computing enables applications to continue to function even when the internet connection is unavailable.
–Customization. Edge computing allows for the customization and personalization of services based on local data.
Despite many benefits, edge computing is not meant to supplant cloud computing. There are major advantages of cloud that edge cannot compete, such as computational power, storage capacity, mobile device connectivity. Instead, designing a smart data infrastructure should take advantage of both edge and cloud and allocate where and how to store, compute, or move each piece of data efficiently.
Edge Computing FAQs
Here are 5 things to know about edge computing.
Smart without cloud. Despite what many told you that smart applications must come with cloud and the Internet, it is not always true (technically). Imagine a smart greenhouse that employs sensor data to automatically control fertilization and irrigation. The central control module can be deployed in a local computing hub within the same LAN as the sensors and the devices. No Internet connectivity is required and the data is more secure, traversing only within the local network. However, if the farmer requires remote monitoring/control of his greenhouse, then cloud or central server is necessary for remote communication.
- Smart without cloud. Despite what many told you that smart applications must come with cloud and the Internet, it is not always true (technically). Imagine a smart greenhouse that employs sensor data to automatically control fertilization and irrigation. The central control module can be deployed in a local computing hub within the same LAN as the sensors and the devices. No Internet connectivity is required and the data is more secure, traversing only within the local network. However, if the farmer requires remote monitoring/control of his greenhouse, then cloud or central server is necessary for remote communication.
- Hardware advancement. Any so-called “smart” device can leverage the edge computing and become smarter. Take voice assistive devices like Amazon Echo. Up until recently, Echo had only processed specific wake words and transmitted the subsequent voice stream to the cloud to process. Now that recent generation of processors is becoming more and more powerful, smart devices, including Echo, are shifting more computation to the edge. Not only this is more optimal from a performance standpoint, but it also allows more customization service opportunity for end users when the device can employ local AI model to learn and develop further functionalities that cater to its owner’s preference.
- Edge safety. While computing data at the edge can keep your data from Internet exposure, security issues might arise from elsewhere, especially physical security. Hacking into a protective environment such as a data center requires some skill. However, stealing data at the edge by simply copying the data or removing the edge device’s disk itself is quite easy if there is no proper safeguarding the physical device against such tampering.
- AI at the edge. Edge computing is often coupled with artificial intelligence (AI) technology, enabling intelligent decision-making at the edge. This integration facilitates real time data analysis, predictive maintenance, and personalized user experiences without relying on constant connectivity to centralized servers.
- 5G. Having 5G connectivity doesn’t necessarily render edge computing unnecessary, as both technologies serve different purposes and complement each other. Despite the high speeds of 5G, there can still be latency in transmitting data to and from centralized servers or the cloud. Edge computing can reduce this latency significantly, which is critical for applications like industrial automation and smart maintenance.
