Data is exploding in every respect: in data volume, data velocity, and data variety (the 3 v’s). One driver of this phenomenon is the growing number of Mobile and IoT devices and thus, data sources. Making this data useful is one of the driving forces behind the adoption of Edge Computing. New use cases don’t only rely on using this data, but also upon the usability and speed of usability of this ever growing data. There are several practical challenges with this growing data volume that drive the adoption of Edge Computing:
New Use Cases Drive Edge Computing
Bandwidth Limitations
The existing network infrastructure cannot support sending all the data to the cloud. Particularly in urban areas there is a concentration of devices and data overburdens existing infrastructure. While 5G promises some relief, it is no hailbringer. First of all, if you want to implement your IoT project now, 5G is not yet available and many questions about 5G remain, e.g. pricing. But moreover, as the number of devices and data is growing ever faster, it is already clear that data volumes will outpace what 5G can support. Edge Computing will be an important technology alongside 5G to enable IoT.
Fast Data Requirements
Response time requirements are growing at the same time as data volumes are increasing. Sending data to the cloud for computation and storage means applications’ response times have a higher latency and depend on the network, which cannot guarantee response rates. Edge computing guarentees significantly faster data. Use cases that need speedy response times or guaranteed responses cannot rely on cloud computing. For example, in driver assistance, where every millisecond counts or in factory floors, where downtimes are too costly.
Sustainability
Sending data to the cloud and storing it there is inefficient and therefore costly – not only in plain €, but with regards to CO2 emissions too. The distance the data needs to travel needs hardware, connectivity and electric power. Therefore, sending data unnecessarily back and forth is wasteful beaviour and burdens the environment unnecessarily. With growing data volumes, that burden is growing. In fact, analysts predict that cloud computing data centers will consume as much as 21% of the total global energy by 2030. [1]
To scale your prototype, you need to move to the edge
At the start of IoT projects, quick prototyping, testing and piloting on early iterations of an application’s functionalities, can effectively be done in the cloud. However, in productive environments when applications scale it is often hard or impossible to keep cloud costs at scale, making the business not viable. Then it is time to move to the edge.
At the same time, decreasing hardware costs and hardware sizes are enabling more complex local computing, meaning there is less need for additional cloud usage. E.g. increasingly AI and ML is done at the edge, including model training.
Data accessibility and Smart Syncing
Today’s successful businesses require a smarter approach to data management and integration. Data synchronization increases operational efficiencies, saving time and resources by eliminating redundant data transfer. With data synchronization, only predefined, useful parts of a data set are sent to a central instance. This means that while large volumes of data can be collected and analyzed locally, not all of this data is sent to and saved in the cloud. This reduces the impact on bandwidth, utilizes the local hardware resources for fast guaranteed response times, and keeps project cloud costs low – ultimately creating a more sustainable and efficient model of data architecture, enabling long term project scalability.
ObjectBox’ current database technology is enabling companies to persist and use data on edge devices, faster than any alternative on the market. It enables networks of edge devices working without a central instance, enabling even more new use cases.