Cloud computing is a new form of outsourcing. Users no longer operate hardware and software components in their own datacentres, but procure these as services via the Internet from an external provider that offers those services to other customers also.
Cloud customers pay only for the exact amount of time that they use or receive the requested services. They use these services based on their current need and can easily increase or decrease the scope and capacity in real time.
Often, cloud computing is only associated with the provision of applications (e.g. Software as a Service: SaaS). However, SaaS is just one part of what cloud computing is all about; in fact, it also includes the provision of virtual machines or memory modules via the Internet. Users can access these computers and storage services without having to face the difficult task of installing or configuring the necessary hardware by themselves.
In principle, all components that businesses operate in their own datacentres can be provided by an outsourcer as a service, as long as they can be delivered via the internet efficiently. Therefore, standard cloud computing packages include, for example:
• Disk space
• Network capacities
• Development tools
Today the technologies that provide a platform for cloud computing have become so reliable that businesses are increasingly taking advantage of new internet services. At the technological core of these services lies virtualisation - a concept familiar since the days of mainframe computing.
It provides users with an abstraction layer that insulates them from the actual physical hardware, such as computers and disk space. Thus, end users perceive the cloud services they receive as if they are obtained from their very own in-house datacentre.
Even though businesses generally have their own well-developed IT infrastructure and hardware, they contract external partners to supply more virtual servers because they want to be prepared when dealing with peak loads.
Often, end users tend to access certain applications only at certain times; for example at the end of the month. IT departments need to be prepared ahead of time for these scheduled peak loads and have corresponding machine capacities available to offset them.
Most of the time, IT resource estimates are calculated like this: the IT department reserves only as many hardware resources as necessary to guarantee standard business operations. If the demand goes up, then it adds virtual computers on a lease basis, which is a more cost-effective method than buying new computers that would otherwise sit idle except for a few days per month.
In addition, cloud computing also helps to expand existing services faster and more flexibly, thereby helping to achieve a greater competitive advantage.
For example, a small research department in a pharmaceutical enterprise might have to triple its computer capacity for a short six-week period in order to accelerate the testing of a new product, allowing it to come to market faster. Thanks to cloud computing, the IT department of such an enterprise could easily provide an additional 20 servers within minutes.
As a result, the enterprise spares itself the not-inconsiderable effort and expense of having to order the servers from the manufacturer, as well as getting them delivered and installed, which would normally take days.
There are other advantages to cloud computing. Customers do not have to worry about updates and migrations, which are now handled by the cloud solution partner. In addition, customers can take advantage of services irrespective of their whereabouts and no longer find themselves bound to a single physical installation location. As a result, enterprises become ultimately more flexible and can move to support their important business processes faster.
It is imperative that users and providers plan and prepare carefully prior to implementing cloud services. Initially, they will have to determine whether cloud services are even legally authorised for the area in question.
They will also have to understand and assess the risks that are naturally present. It is crucial for firms to analyse how business-critical the services are that will be outsourced to the cloud. This is because business-critical services require a high degree of system stability, a high level of service quality, and redundant service capacities built into the system.
Cloud customers usually see a decrease in IT management costs because part - and only a part - of management is outsourced. However, the cloud is an area that still has to be monitored and measured. If customers want to integrate cloud services seamlessly into their enterprise, cloud service management has to be integrated tightly into their own IT infrastructure management.
Thus, cloud users need all the management tools that they require to manage their own IT anyway. The tools also have to be able to manage virtual and physical environments at the same time.
To the IT infrastructure, virtualisation adds another layer that management software has to identify, integrate, and manage accordingly. In addition, users need to utilise a balanced workload automation management tool to ensure that some of the workload is assigned to virtual machines when necessary. In addition, they should also check whether the virtual hardware does ultimately cost less in terms of time and money than choosing to operate a physical equivalent in their own datacentres.
If aspects of IT infrastructure management are transferred to a certain extent from the IT department to the cloud, then the management expenses of cloud providers increases accordingly.
The more services they provide, the more important - and more complex - it becomes to manage their own IT infrastructure. However, this management function is a load-bearing pillar of their core business. For that reason, they require highly integrated solutions that unify and simplify management.
Basic requirements of such solutions include, for example, the centralised control of all services in distributed and virtual environments, as well as the seamless integration of all management tasks to obtain an overview of the performance and availability of the services offered.
Should individual components fail, tools are required to quickly find the root cause and correct the faults contained therein. In order to get a handle on the increasing complexity of the IT infrastructure and to ensure smooth operation of the cloud services, management tools must be able to offer a high degree of automation.
It takes an efficient infrastructure management system for cloud providers to ensure and demonstrate that they deliver services at the service level agreed upon at the beginning, that data and applications are secure, that legal provisions and guidelines are complied with, and that changes can dynamically be carried out.
Michael Jaeger is the senior principal product marketing manager at CA.For all the latest tech news from the UAE and Gulf countries, follow us on Twitter and Linkedin, like us on Facebook and subscribe to our YouTube page, which is updated daily.
Subscribe to Arabian Business' newsletter to receive the latest breaking news and business stories in Dubai,the UAE and the GCC straight to your inbox.