The use of cloud technologies to deliver applications and data to customers at multiple locations.
It can sometimes be necessary to have your cloud services in specific locations — perhaps for performance or regulatory reasons, for instance GDPR legislation.
Distributed cloud computing ensures the complete management and optimal placement of data, compute and network resources to meet your customers’ needs.
What is it?
For most of its history, cloud has been about moving applications from physical data centers into a centralized cloud system. But distributed cloud recognizes that there can be benefits to having cloud-based applications and data sources at specific locations. This is often for performance or regulatory reasons.
In distributed cloud computing, your cloud provider maintains responsibility for the operation, governance and updates of the solution, while providing services at specific locations, rather than a centralized hub. This can allow you to improve performance, if you have time-critical applications; or you may be required to store customer data in specific locations.
What’s in for you?
Distributed cloud computing enables you to take advantage of the dynamic scaling, self-serve capabilities and on-demand computing that makes cloud so compelling. But it also adds the ability to move compute power and storage to where you need it.
That’s a significant benefit if you run applications where latency or throughput are business-critical. It’s also useful if your enterprise is covered by regulations, such as the European Union’s data laws, that require you to store and process data in specific locations.
What are the trade offs?
Migrating to the cloud requires enterprises to adjust to a new level of complexity that is inherent in distributed systems. It’s somewhat straightforward to troubleshoot issues that arise in applications running in physical data centers; when those applications are distributed across various virtual systems inside the cloud, management becomes a little trickier. With distributed cloud, those management issues are increased further.
Many of today’s enterprises lack the maturity to operate effectively in a distributed cloud because much of their legacy estate was not designed to operate in the cloud. Unless they have been willing to invest the time and money in rearchitecting and replatforming their applications, they can expect to hit a few bumps in the road to distributed cloud.
How is it being used?
Some of the earliest applications of distributed cloud were content delivery networks — think Netflix, Spotify and the like — where having ‘mini clouds’ in specific geographic locations enabled the reliable delivery of content for better performance and user experience without lag or buffering.
It’s also being used for applications where latency is critical — in areas such as banks’ trading systems or autonomous vehicle solutions.
Increasingly, we see enterprises that are deploying Internet of Things solutions — where thousands of remote, connected sensors are used to gain operational insights — investigating distributed cloud computing. Here, the intention is to do some of the necessary data processing close to the data sources, so that decisions can be made in real time.