Edge computing is a term that’s regularly coming up in technology conversations, and being touted as the next big thing after cloud computing. There might be a great deal of truth to that given the fact that a MarketsandMarkets report forecasts edge computing to grow at a CAGR of 35%, and reach $6.72 billion USD by 2022.
So, what exactly is edge computing?
Research firm IDC describes edge computing as a “mesh network of micro data centers that process or store critical data locally and push all received data to a central data center or cloud storage repository, in a footprint of less than 100 square feet”.
Basically, it’s processing data at the edge of a network, at or near the point of origin. This ‘micro data center’ could be a sensor device itself or a device with predefined computational powers that will locally process time-sensitive data and relay it back into the system. The rest of the data is then moved to the cloud for further processing.
How is it better than cloud computing?
Edge computing brings in certain distinct benefits over cloud computing, but that requires us to first acknowledge the rising importance of IoT across industries.
IoT demands a complete ecosystem of connected sensors and devices that are continuously gathering a massive volume of data. Giving this data a round-trip to the cloud/central data centre is slow and costly. What follows is where edge computing has an ‘edge’ over cloud computing.
Sending and receiving data from the cloud, especially when data centres are physically located miles apart, can be slow. Not for us, but definitely slow for enterprises whose businesses run on the speed of data processing.
With IoT, real-time data processing has become a requirement. Sensors monitoring manufacturing lines, or cars navigating via GPS and sensor data, need to process information in real-time to be able to correctly respond to situations. And hence, waiting for the data to get back from the cloud is not a feasible option.
With edge computing, critical data is processed near to these IoT devices. This eliminates latency issues and vastly improves response rates across enterprise operations.
It’s predicted that by 2020, there will be over 50 billion IoT devices collecting more than 1.44 billion data points per plant, per day. First up, that is a massive amount of data to be transferred over the network, leading to increased loads. Secondly, even if it’s done, it will be hugely expensive for businesses. There will be the cost of acquiring additional bandwidth to transfer this volume of data. Add to that the fact that they have to increase investments in load balancing and frequent maintenance of the network and data centres.
With edge computing, on the other hand, the majority of the time-sensitive data is processed locally, and relayed back to the IoT devices for further action. That leaves a manageable amount of data that needs to be transferred to the cloud, and hence does not demand huge expenditure.
With data travelling long distances to reach the cloud, there are high chances of data getting corrupted. This can cause data loss, system crashes, and also financial loss due to incorrect data processing.
In the case of edge computing, the path from an IoT device to micro data centre is extremely short. This ensures that there is a slim-to-none chance of data corruption, or network jitters to cause data packets to reach unevenly. The data transmission is reliable and hence the data processing and insights are also more reliable.
Edge computing gives businesses the option of not just locally processing, but also locally storing sensitive data. This is definitely a step beyond having to store data in public or hybrid clouds where it is slightly more prone to security risks.
Because edge computing capabilities are always close to the data source, they are almost always within premises controlled by the enterprises themselves. This means they can build and deploy custom security measures as per their standards of compliance.
Will edge computing replace cloud computing?
While edge computing definitely has its upside, one must remember that it’s a network of ‘micro’ data centres. They are great for processing time-sensitive data or storing critical data; but that is only a portion of the data being produced by enterprises. There’s still a huge amount of data that needs to be stored and processed at normal speeds. And there’s nothing better than cloud computing to achieve that.
So no, cloud computing will definitely not be replaced by edge computing. What will happen, though, is that both these methods will become indispensable and complementary parts of an enterprise data strategy. So it is advisable for businesses to start understanding and investing in building edge computing capabilities, so they get a headstart in the game.
Author Bio Sriram Sitaraman: Practice Head for Analytics and Data Science at Srijan Technologies. With over 20 years of experience in designing and delivering innovative business solutions, Sriram leverages his expertise in machine learning, statistical modelling, and business intelligence to enable digital transformation in industries as diverse as healthcare, manufacturing, retail, banking, and more.