Scott Galvin, Executive Vice President, Technical Solutions
I hadn’t realized exactly how firmly the virtualization buzzword had taken hold of your average banker until a recent conversation with a prospective client. This particular financial institution had managed to exist through the years without any network to speak of, but now faced the challenge of needing to move into real-time processing, and setting up a basic network was in order. As I went through my usual discovery questions to determine this institution’s wants and needs that their network would have to support, the customer interrupted me and said, “I only know two things about networks. I know that everyone needs their own passwords and I know that we need to virtualize.” It was with that statement that I came to realize the idea of virtual networking has moved out of the realm of the highly technical administrators and CIOs and has been firmly implanted in the minds of the CFO or loan officer who was lucky enough to inherit the responsibility of IT in their institution along with their traditional duties. My other thought was with so many technical buzzwords, the details of how, when, and why you use virtualization are still a bit unclear to the both the people who deal with technology part time and even many who make it their sole focus. With these two ideas in mind, I would like to take a few minutes to discuss what virtualization is and how you can use it to actually improve your institution’s ability to service your customers.
First off, what is virtualization? Virtualization is a method of converting a traditional set of physical resources to work on a versatile platform that can support similar technology with greater scalability and efficiency. Think of virtualization as a giant multiplex movie theater where each of the screens or individual theaters equates to a virtual machine. In creating each individual theater, they decided how much space, seating, movie equipment, etc. is going to be allocated, much the same as you divide up the resources (i.e. RAM, CPU, disk space) to run each virtual server. It’s also similar in the way that a movie can be showing on a screen or the screen can be dark just as servers can easily be brought up and down without affecting the other servers in virtual format, despite the fact that they exist on the same physical host. Finally, if the movie company were to run into trouble showing the movie on one particular screen, it could be easily transferred onto another screen with minimal issues which is very similar to how if a box is having an issue running a virtual server, that virtual server can easily be transferred to run on another host. Virtualization is essentially put in place to increase your options regarding how you manage your network and can happen at several different layers. You can virtualize hardware, operating systems, and even specific applications. For the sake of this article, we’ll focus on hardware virtualization.
Now that we have at least a rough idea of what virtualization is, the next question is why should we use it? Previously it was said that virtualization gives me more options on how I want to manage my network. Many of you are probably thinking, “Why would I want more choices? I have too many choices as it is!” While the choices at times can appear overwhelming, in many ways the lack of options provided by utilizing a traditional physical infrastructure can limit your ability to grow your network, and limit the speed with which you can recover your infrastructure from both large disasters or even small disruptions such as a crashed server. Let’s tackle scalability first. Your network can be required to grow to meet many different types of demands. For example, if your institution decides to add a service, your network will likely require a new server to house the software and provide resources to support the service. It’s not uncommon for many institutions that adding a server can take from a few weeks to a month or more. Another way demand on your network can grow is if your user base expands significantly. In that instance, it may turn out that one or more of your application servers just do not have enough horsepower to accommodate for the new load. To remedy this situation, the IT staff could be forced into replacing or adding multiple servers just to increase the capacity. A project of this scope can equate to something just short of a network overhaul or major project and drain your IT staff’s time and resources. Virtualization can significantly reduce the burden in either of these circumstances. For a new application, your IT staff could simply create a new virtual server on your existing infrastructure out of a preconfigured server template built to meet all requirements set by internal policies and best practices. Using this template allows a brand new server to be brought up in minutes and is instantly ready for installation and configuration of the new software, reducing a wait of weeks and months into a few days of planning and implementation. As for the capacity scenario, virtualization allows you the option to increase the RAM, processor and drive space with just a few clicks of the mouse, turning what traditionally was a major project to be resolved in minutes with minimal effort.
The final aspect I would like to cover is how virtualization can aid you in the areas of redundancy and recoverability. Whether you are planning how to maximize continuity during something as minor as a server crash or major such as a hurricane or tornado, virtualization can play a important role. In a high-end feature rich configuration, virtual servers can be configured to automatically failover to properly functioning hardware when the node that was originally hosting the VM suffers a hardware event. This kind of automatic failover can result in zero downtime and requires no human interaction, meaning that even when events happen in the middle of the night, you are still up and operating. In a less feature rich environment, the same type of recovery can be performed with only minutes of manual interaction. Both of these results are significant improvements over the current options. Today, when your traditional physical server has a hardware issue, the likely scenario requires your IT team spending half to a full day restoring files from your backup and working through application configuration and permission issues. With regard to the true disaster recovery side, virtualization can be implemented to not only eliminate the need for the institution to have duplicates of each physical server they require to conduct business, but can also significantly reduce the recovery time of those critical servers whether in a real disaster or merely helping streamline your annual testing process.
Hopefully this general idea of what virtualization is and why you would implement it has helped clear away some of the mystery surrounding this buzzword. As this technology is becoming more commonplace, we want to continue to distribute information and ideas to build the best infrastructure for each financial institution.