Understanding the role and opportunities of edge computing
“The most profound technologies are those that disappear. They weave themselves into the fabric of everyday life until they are indistinguishable from it.”
When Mark Weisler, then chief technologist at Xerox PARC, uttered these words in reference to the Internet of Things (IoT) as a concept back in the 1990s, he probably had no idea at the time what an accurate prediction it would turn out to be.
Fast forward to the Information Age and Weisler’s words have shifted from concept to reality. The proliferation of IoT is such that it’s expected there will be nearly 30.73 billion IoT connected devices by next year.
For cloud and associated technologies in their current state, the sheer volume of data being generated and the related need for speed is starting to present a challenge in terms of converting data quickly and accurately into useful insights. But, where there’s challenge there’s almost always opportunity, and in this context, solutions that support data processing either at or near the source of where it is generated are coming to the fore.
IoT meets edge computing
Enter edge computing, the practice of processing data at or near the source of the data, as opposed to relaying it to a centralized location, as is done with cloud computing.
The “edge” refers to the actual space of the distribution, near the edge of the network. In edge computing, data is processed by the device itself or by a local computer or server, which eliminates costs and ensures that applications can be used effectively in remote locations. While some data will always need to be processed in an actual data center, companies can minimize Internet bandwidth usage while boosting performance and reliability by incorporating edge locations.
The need for edge computing is growing
While edge computing is not necessarily new – remote working and branch offices have been a vital part of business practices for years and require computing in multiple locations – we are seeing a major increase in the edge data center market because of the dramatic increase in device usage and subsequent data generation.
Edge computing can be ideal for IoT for a few reasons. First, the fact that the data is processed much closer to the source means that latency between devices and data processing layers is reduced, which equates to faster response and better decision making. Costs are also lowered as less data is required to be processed in the cloud, and network capacity for other workloads increases.
Second, because edge computing enables the data to be localized, if a device does fail there is no ripple effect on other devices. Keeping the data locally also helps to heighten security and compliance as fewer opportunities are presented for cyber criminals to access all data simultaneously.
Last, delivering and keeping data in smaller repositories makes it far easier to compartmentalize data according to type and region for example, rather than having to pull extracts of data from one central database. It makes for easier aggregation of data as well as the provision of analytics in real-time.
Incorporating edge computing into your data center strategy
As we continue to incorporate more devices into our everyday lives and businesses, the ‘data monster’ will keep on growing. It’s estimated that by 2020, 1.7MB of data will be created every second for every person on earth.
There’s no going back, therefore it’s important for data centers to adapt and provision for this growth. It is not an easy fix or an instant adjustment - executing an edge computing strategy will take time and careful planning, as well as resources and geographical readiness.
Don’t put all the focus on the short-term goals. When planning, organizations need to ask themselves where they are now, where they will be in five years and what will they do to enable change?
Some key features that should be considered and planned for in an edge computing strategy include:
- Easy deployment – as the compute is often in more remote locations without full IT staff to attend to it, any edge computing strategy needs to be easy and quick to install and get up and running
- Reliability – businesses that rely on edge computing can have hundreds of sites and cannot afford to constantly be traveling to check in on them, so systems need to be reliable and have longevity
- Automation – due to location, automation is key to keep the systems running without IT too much manual attention
- Self-Healing – edge computing systems must run with as little management as possible with any tasks needed being performed remotely and with ease
- Scalablity – edge computing needs often vary depending on the type of business, so these systems should be able to scale up and down again
Edge computing is also primed to play a key part in the continuing deployment of IoT devices. The online gaming and gambling industry for example will also start to drive a different type of IoT market as controllers strive to provide even more realistic experiences. As IoT devices continue to insert themselves into all areas of life, the demand for edge computing will become much stronger.