BACKGROUND IMAGE: iSTOCK/GETTY IMAGES
If you think of convergence as getting to use the virtual server technology of your choice, but with a pre-designed infrastructure combination that you can buy behind it, the public cloud becomes a further extension of that same concept. With a cloud storage architecture, you decide on the type and configuration of virtual servers required to solve a given business problem and then simply gain access to that virtual server via a network connection to the cloud provider. While you give up the ability to build, customize, or even see or touch the infrastructure, you get a pre-designed architecture that the service provider worries about.
Like converged systems vendors, cloud providers give you a list of choices for the infrastructure and the amount of RAM or disk space associated with it. However, you do not get to choose which vendors' components go into the offering or where it specifically resides within the remote data center.
Convergence can radically simplify the process of choosing infrastructure and building the technology, but a cloud storage architecture can go further by allowing users to do the following:
Use trial and error to find the right approach. When you're building a storage architecture, it's hard to hit the nail on the head every time, especially if the requirements aren't clearly outlined. With a traditional or converged infrastructure, you still own gear for three to five years. You may be able to repurpose that gear for another offering (if it's a good fit for what you purchased), but it's a tricky proposition and the hardware you bought but didn't like may end up wasted. A cloud-based storage architecture allows you to build out an offering and, if it's not quite right, simply shut off the initial try and get something else.
Scale up and down to meet demand spikes. Many business situations create spikey demand for IT. Whether it's retail demand around Christmas, processing for a month-end close or just the variability from daytime business hours to nighttime slowdown, demand increases and decreases continually. With traditional or converged products, you need to buy an infrastructure scalable enough to run as many virtual servers as you will ever need to run. With a cloud architecture, you can scale up or down to meet demand as it comes, and pay less when you use fewer resources.
Set up the environment quickly. Converged systems can reduce the lead time of buying from many suppliers and the effort of putting it all together. But in the end, you're still ordering, standing up and deploying products when the demand arises. Most cloud providers have systems ready and waiting -- especially in the case of frequently used storage types -- and they can measure turnaround in minutes or hours, rather than days or weeks.
Designing a cloud-ready infrastructure
Delving into the difference between virtualization and the cloud