Why Edge Computing?
Increasingly companies are wanting to use data streaming from the edge to better understand their business. However, there are technical challenges (such as latency and bandwidth), with transmitting these ever-increasing data volumes from the edge into a centralized data centre for computing. Edge computing is a distributed computing paradigm and an architectural pattern whose goal is to shift the compute layer closer to the data sources to address these challenges. Moreover, the unprecedented scale of data that is being created by connected devices has outpaced network and infrastructure capabilities that are needed to push this data into the cloud for compute and analysis. The edge computing paradigm offers an alternative and effective solution to this problem. The principle of edge computing is simple: If you can’t get the data closer to the data centre in time and at the speed you want, get the data centre closer to the data. The concept of edge computing isn’t new though. It is rooted in decades-old ideas of remote computing (think remote offices and branch offices). However, the use case that edge computing addresses have evolved.
Imagine a scenario where a self-driving car had to feed data from its plethora of sensors up to the cloud and wait for a response before it carried out a manoeuvre. It’s not difficult to see how disastrous that would be with the inconsistency of the cellular network that connect the car to the cloud. In the words of the hard rock band Aerosmith, we’d be – “Living on the edge”. However, self-driving cars are a reality today, a reality made possible by edge computing.
How does Edge Computing help?
Edge computing offers a more efficient alternative by bringing enterprise applications closer to data sources and thereby delivering substantial benefits such as faster insights, improved response times and optimised bandwidth consumption and mitigation of network disruptions. With edge computing data is processed and analysed closer to the point where it is created thereby enabling faster insights and improved customer experiences. By increasing the speed at which value from data can be retrieved, edge computing increases the effectiveness at which businesses can respond to insights in data. Edge computing is transmuting the way data is acquired, analysed, and actioned. In edge computing only the result of the computing work at the edge, such as real-time insights, maintenance predictions and other actionable insights, is sent back to the main data centre for stakeholder review and action.
Edge Computing Challenges
Edge computing solitons are composed of heterogeneous components such as compute systems, connectivity, and storage, which if not appropriately managed can soon become a logistical nightmare. Edge computing might not be as secure as a centralized or cloud-based system, making it a weak point for potential security vulnerabilities. Manging each remote Edge location requires appropriate processes and monitor making it difficult to visualize and understand the state of the entire Edge environment. Hardware and upgrade failures can happen. NICs can fail, power supplies can burn out and are challenging to fix especially if the deployed edge is in a remote location. A fail-proof approach to push out system configuration changes, software or security updates, and to deploy code updates to models or algorithms is needed in order to ensure critical business processes supported by the edge compute ecosystem are not impacted.
Use Cases for Edge Computing
Oil and gas industry
Services that the oil and gas industry delivers is critical requiring their assets to be carefully monitored. However, oil and gas plants are often in remote locations and edge computing reduces dependency on good quality connectivity to a centralised data centre.
Healthcare
The healthcare industry has several use cases where edge computing can be of substantial benefit. An edge at the hospital site that processes and deletes data locally can aid in meeting data privacy laws. Edge also enables near time notifications to health care professionals on site of unusual patient trends picked up by wearable monitors.
Content delivery
Caching content such as video stream and web pages at the edge improves user experience greatly and significantly reduce latency. Content providers rely heavily on global edge farms to meet the ever-increasing demand for streaming services where the speed of content delivery can be a matter of survival in a very competitive marketplace.
Conclusion
Business use cases that need low latency, real-time and near real-time processing, or a series of back-and-forth operations, has displaced the role that the cloud played in centrally processing data necessitating a change in topology for where compute power resides. Gartner estimates that by 2025, 75% of data will be processed outside the traditional data centre/cloud up from 10% today. Edge computing is here to stay. In fact, IDC estimates the number of connected devices will reach 55.7 billion by 2025. The more connected devices the more will be the probability for use cases that require edge computing.
