Thinking of Infrastructure as a Service, chances are that you would think of Rackspace, Verizon Cloud or Amazon Web Services. Where Microsoft Azure and Google Compute Engine are relatively new entrants, Amazon Web Services has been around with the likes of Rackspace. Opting to work with one provider over another is based on a myriad of factors of which many might be decided by the developer. The basic idea of Infrastructure as a Service is that you configure virtual machines, deploy and run them virtually from a set of dispersed locations, but each provider offers many options from storage to compute to database to development options. Knowing your development and hosting requirements will simplify choices greatly. Some providers also charge less for advance reservation of resources, which might work out cheaper that the pay-as-you-go model. For example, Rackspace offers close to 20 % discount for advance reservation for a period of 6, 12, 18 or 24 months.
The following are the main criteria to bear in mind when choosing an Infrastructure as a Service provider ( IAAS Providers ):
- Infrastructure: A robust, managed IaaS cloud solution running in an highly secure data center with reliable cloud services. The provider must have a 24/7 automated monitoring and management practice in place. Understand and explore the capability of the provider to offer flexible capacity to accommodate variations, through a vendor like Nalashaa.
- Architecting a cloud solution: A partner must be able to give a consultative approach to build the best suited cloud solution for the requirement presented. Picking a pre-packed solution will invariably pose problems in the long run, when you would need to scale up or down due to any circumstance, or can result in opting for a expensive solution, to ensure flexibility. More features and functionality being offered with an option to design your solution has maximum appeal, and to build your solution, take assistance from an experienced partner
- Compute options: Prebuilt services such as the database you would need, if available with your Infrastructure as a Service provider, will make the process of setting up and maintenance simpler. An auto scaling function and load balancer is a feature which works great for businesses which are erratic in nature and customer centric. A load balancer is responsible for distributing incoming traffic among servers hosting the same application content. By balancing application requests across multiple servers, a load balancer prevents any application server from becoming a single point of failure, thus improving overall application availability and responsiveness. Another point to bear in mind is the use of some particular compute service, such as Hadoop, then having the provider host this would mean less work for you to set up the system, and getting your work done more quickly.
- Application Development: If you have a language of preference for your application development needs, this also makes your choice of provider for Infrastructure as a Service simpler. Microsoft, Google, Joyent etc have some specific languages which they work with, such as Microsoft Azure works well for .Net applications. It must be added that many providers also add new app development services frequently to lure in more customers.
Apart from these criteria, pricing and security are two other factors on which many cloud solution decisions are based on. Many customers feel a strong lure towards Google’s highly secure infrastructure offering, where as others are drawn by the varied offering from Amazon Web Services. Then there are also many who opt for the hybrid approach put forth by Azure, HP, Openstack, Rackspace etc.
Nalashaa has had varied experience in setting up and managing Infrastructure as a Service offerings for many customers across various providers. We also have a deep understanding and working knowledge of various provider offerings and offer assistance in understanding specific client software requirements and helping you pick an appropriate provider.
Latest posts by Aathira Nair (see all)
- Analytics for Healthcare – 3 steps on getting started - May 6, 2016
- How does Technology stand up to the Data driven Healthcare model? - September 28, 2015
- ICD-10 – Bringing in practicality to the anxiety - July 23, 2015