Why Edge Computing? Traditionally, all data from your infrastructure was sent to the cloud for analysis. With the advent of the internet of things, this paradigm is changing and the 2 main drivers are time and cost
- Costs: With billions of connected devices transmitting data- it is neither efficient nor economical to send all that data directly to the cloud using expensive bandwidth and storage.
- Time : Another key factor is the time sensitivity of a lot of this data. Much of the data generated by things is perishable and has value only for a short period of time and needs to be analyzed quickly to gain any insights or take any useful action.
The nature of the data and its time sensitive usefulness demands that some processing happens at the edge. This is definitely a growing trend and as Gartner predicts by 2022, 75% of enterprise generated data will be created and processed outside the data center or cloud, up from just 10% today.
Real Time Decision Making for Connected Cars :Let’s take the smart car example- the decision of whether an airbag should be deployed or not is not something that should happen in the cloud. It needs to happen at the edge, where the car makes this decision in real time based on data from cameras, sensors, vehicle to vehicle communication details, weather information etc. For this to happen we need edge devices to have compute and storage capabilities to power edge analytics which the majority of things and sensors like smart cameras, smart lights, smart meters etc. do not have. This is where hyper-convergence at the IoT Edge can be a game changer and allow industries to move more and more processing towards the edge for better control over their data, faster decision making and of course lowering bandwidth and storage costs.
Hyperconvergence At The Edge: By providing compute, storage and network capabilities to these little micro-data centers you can fully realize the benefits of a fully distributed IoT architecture where your analysis is split between the cloud and the edge based on your use cases and needs. You also get to enjoy the same benefits of virtualization you see in the cloud by lowering your total cost of ownership and flexibility to scale. And most importantly you will be able to have a consistent data center strategy from the edge to the core to the cloud. Like it or not the datacenter is moving to the edges of your network- be it a factory, a hospital or a jumbo jet and hyper-convergence at the edge is the new IoT model for success.
Stay tuned for the next post where we will discuss what factors are driving edge computing requirements.