How the edge and the cloud tackle latency, security and bandwidth issuesMark Seddon, CEO of Pact Global, discusses how edge computing is complementing the cloud to tackle latency, security and bandwidth issues
Edge computing, an IT deployment designed to put applications and data as close as possible to the users or ‘things’ that need them, is best understood through its use in the Internet of things (IoT), as it is the IoT that has produced the need for it. Simply put, IoT is all the physical objects that connect to the internet and exchange data; thermostats, security cameras, fridge freezers, Alexa, Google Home and even vehicles. Just as the need for increased data storage by individuals and companies created the need for the vast centralised storage capabilities of the cloud, IoT has created a need for a faster, more secure way to use the same data, but by using less bandwidth.
The move from personal computing to cloud computing has seen massive amounts of data sent to and stored in huge server farms, largely owned by Google, Amazon, Microsoft and IBM. In order to use cloud data, it must be accessed, processed and analysed before being returned for purpose. A useful analogy for this is the home assistant. When you ask Google Home what the weather is going to be like, it processes your speech, sends a compressed version to the cloud, which is uncompressed, processed, perhaps performs an API function to get the answer, and returns it to your device. This round-trip data usage creates three main issues: latency, security and bandwidth.
How to control access to IoT data
With the rise of IoT, edge computing is rapidly gaining popularity as it solves the issues the IoT has when interacting with the cloud. If you picture all your smart devices in a circle, the cloud is centralised in the middle of them; edge computing happens on the edge of that cloud. Literally referring to geographic location, edge computing happens much nearer a device or business, whatever ‘thing’ is transmitting the data. These computing resources are decentralised from data centres; they are on the ‘edge’, and it is here that the data gets processed. With edge computing, data is scrutinised and analysed at the site of production, with only relevant data being sent to the cloud for storage. This means much less data is being sent to the cloud, reducing bandwidth use, privacy and security breaches are more likely at the site of the device making ‘hacking’ a device much harder, and the speed of interaction with data increases dramatically.
While edge and cloud computing are often seen as mutually exclusive approaches, larger IoT projects frequently require a combination of both. Take driverless cars as an example. If information from all car sensors had to be sent to the cloud for processing and returned for function, the capability of the network, the ability to hack into the car and the delay in response would all mean self-driving cars were not feasible. As a combination of cloud and edge computing, there is no user responsibility for the software that is run and sensors work in real-time, but the car has to use centralised data to get updates and send processed data back to enhance algorithms.
Q&A: CTO tips on delivering cloud innovation to avoid disruption
Despite its maturity in the market, businesses are only recently realising how IoT will help automate and refine the services they provide. Edge computing and the cloud are beneficial to business in a number of ways: time saved analysing data; downsized storage volumes; and ease in abiding security and data privacy regulations like the General Data Protection Regulation (GDPR) being just a few. As the digital and ‘real’ world converge and experiences become increasingly immersive, the proliferation of data being collected is unparalleled and will continue to grow.
Insurance is perfectly positioned to be the first major industry to benefit from the combination of edge computing and the cloud, to create the most immersive user experience with IoT. As an example, home insurance has always been reactive rather than proactive and relied upon users to report a claim to a third party via the phone. Additionally, this data is rarely used to calculate accurate risk. However, with the rise of smart home and IoT devices, home insurers are able to flip this model on its head.
Time is up for the insurance industry to embrace cloud technology
With the use of IoT devices customers’ homes can provide data, which can be leveraged to more comprehensively calculate risk, while smart security and sensors can be used preventatively in the home, sending both homeowners and their insurers alerts when something isn’t right. Flood damage is one of the leading causes of home insurance claims and the perfect example for this. Leak sensors deployed near potential leak areas like boilers, sinks and washing machines, can send notifications to avoid a damaging event. If a pipe bursts when no one is home, the sensor can command a smart valve to shut the water off, minimising damage, notifying the customer and minimising the claim.
For home insurers, partnering with IoT companies can lead to better relationships with customers and improve the understanding of risk, specific to each customer. With this understanding of the true risks it will enable insurers to reduce premiums.
There is no doubt that the future of IoT for insurance is the combination of both edge and cloud computing, utilising the major advantages of each; maximising the benefits for their customers by minimising damage and providing more accurate premiums, whilst minimising risk and claims costs for companies.