The cloud holds many promises for companies, such as the possibility of optimizing costs by paying for computing resources on an "on-demand" basis, improved reliability, lower administration costs, and others still. At the same time, many issues still remain. There is still a lot of confusion related to the security of cloud solutions, both in terms of data security, that is, the capacity of cloud providers to protect their customer's data, as well as in terms of legal security for the stored information, which is the compliance with legislation (many times from multiple countries) and the legal risks associated with it.
While many different areas from IT can benefit from employing the cloud, for many different reasons, there are some use cases where cloud computing fits particularly well. These cases represent situations where the benefits from adopting a cloud solution clearly outweigh the potential risks associated with it, and are the use cases that many companies are choosing to experiment with in the cloud.
The first case where the cloud - mostly cloud servers - is being heavily used is for application testing. Developers have been using virtual machines to simulate different execution environments for quite some time already, but the cloud allows them to do this on a much greater scale. By creating virtual machine images that simulate the final production environment, any developer can quickly spin up a test machine, deploy their projects to them, and run "live" tests on a separate environment, without having to rely on running virtual machines on their own PC, or even on the corporate data center.
The public cloud infrastructure is very interesting in this case because it is a very low-cost alternative. With a few dollars, anyone can spin up servers with multiple hardware / software configurations to test applications under different environments. It also allows controlled testing of large-scale distributed applications, which would be much harder under conventional circumstances.
Since the data being used for application testing isn't (or at least, shouldn't be) real customer data, the risk of data theft or of infringing some legislation regarding where data may or may not be stored is minimal. And since the test servers can be spun up for a test run, and then shut down, companies can greatly benefit from cost optimizations by moving their test environments to the cloud.
Application contingency is another area where the cloud shines. The ability to quickly replicate virtual machine images across different regions creates the possibility for companies to maintain very low-cost contingency infrastructures. If something goes wrong with the main site - a hardware failure, network connectivity problems, even software crashes - new instances of the application servers can be spun up in alternative locations, minimizing application downtime. This means that anyone can have a good contingency plan without much investment in infrastructure, and without having to actually pay for the contingency at all times.
Another possible scenario is to keep multiple servers running in geographically disperse regions. Most companies don't have access to multiple datacenters at different locations internally, and here is where public cloud infrastructure can really help. All the top cloud infrastructure providers offer multiple datacenter locations where servers can be replicated, so it becomes easy to spin up servers in different places. They also offer load balancers and dynamic IPs that can be shared across servers, making contingency that much easier.
Finally, contingency doesn't apply only to disaster cases. Companies can use cloud infrastructure to deploy contingencies for situations where they need more computing power, such as handling peak loads that only happen sporadically. An e-commerce website that has a huge peak of visitors during the Christmas season could run cloud servers to balance out its load only during this time of the year, reducing total cost.
While the risks in this usage scenario are greater, the potential benefits still outweigh them. Most companies can use the cloud to create a much stronger contingency plan than they would be able to relying only on their own infrastructure. There are also several strategies and tools that can be employed to mitigate the associated risks, ranging from data encryption to simply hosting the data separately from the application that is making use of it, on a more secure environment.
"Big" data processing
While there isn't a very clear definition of what exactly constitutes "Big Data" and what doesn't, companies can benefit from employing the cloud to do massive parallel processing of large quantities of data. For organizations that need to crunch massive amounts of data, the cloud brings the possibility to deploy huge numbers of servers for the task. In fact, many cloud providers are already offering servers or services based on the most commonly used tools for Big Data tasks, such as Hadoop.
Even companies that don't have petabytes of data to process every day can benefit from using the cloud for their processing needs. By deploying servers on an on-demand basis, they fine tune their processing jobs with respect to cost vs. execution time. Furthermore, relying on outside infrastructure means a much smaller initial investment in hardware (even if total costs may be higher in the long run), as well as no excess capacity. They can pay only for what they use, and nothing else.
This is the scenario with the largest risks. Companies looking to move processing loads to the cloud must carefully consider the nature of the processing job they want to run as well as the nature of the data being processed, and the risk of this data being stolen. At the same time, the potential benefits are great. For small companies that can't afford to build their own infrastructure, the cloud gives the opportunity to do much more than what was possible before.
As with any other technology, there are places where the cloud shines and offers great benefits and others where that isn't quite true. The scenarios compiled here are the top use cases where I've seen companies employing cloud computing, especially cloud servers. This list, however, is far from complete. It was built from personal experience, without relying on any kind of poll. If you have any other interesting cloud use cases, please share in the comments or let me know.
After working for a database company for 8 years, Thoran Rodrigues took the opportunity to open a cloud services company. For two years his company has been providing services for several of the largest e-commerce companies in Brazil, and over this time he had the opportunity to work on large scale projects ranging from data retrieval to high-availability critical services.