Here is a small business virtualization example in which we consolidate multiple servers down to one:
We have discussed server virtualization in Will Virtualization Cost More Than A Standard Network Installation?, Virtualization And Small Business Benefits - Dollars And Sense and other posts. Small business owners and leaders often ask how all of this can be done on one server. How do they rely on one server for their entire business. Let's address this in a couple of parts:
At a high level, how does server virtualization enable consolidation of servers technically? Say your business has five servers - three terminal servers and two SQL servers. Let's assume that the terminal servers are DL380 G7 E5620 machines with 12GB RAM and roughly 400GB in hard drives. Assume the SQL servers are the same except that they are dual processor, have 16GB RAM and roughly 700GB in hard drives. Because these servers as individual entities will naturally be underutilized, server virtualization allows us to consolidate all of these into one DL380 G7 X5650 with two processors and 40GB RAM with whatever storage we would like (e.g. a 4TB NAS device). By using a more powerful set of processors and more memory, we are able to take advantage of a shared infrastructure to achieve the exact same functions as we would if handled by individual physical servers. As noted in previous blog posts, the economics are compelling. Think of this as the difference between a multi-lane highway and single lane roads. The multi-lane highway is going to allow for more cars to get through because it enables better utilization of the same real estate. Bottlenecks are reduced and higher-speed traffic can take advantage of additional lanes as opposed to waiting in line.
How can this be safe? What if the server goes down, some ask? Recovering from unexpected downtime should always be a consideration in any IT scenario. It is no different with server virtualization. As with many things, there are a range of options. Some customers opt to use the servers that they decommissioned as backup servers. In some cases, they leave them powered down and only power them up when needed. Because recovery using virtualization is so fast, this is a viable option for some businesses. Others prefer a more fault-tolerant architecture and deploy pairs of virtualization servers and a Storage Area Network (SAN) so that the infrastructure runs without interruption. Depending on the needs of the business, there is an appropriate deployment model that will fit the required recovery time. There is always a trade-off between cost and time to recover, but virtualization provides a safe and more agile environment which may be optimized to cost less from an on-going operations perspective.
When one is not familiar with the notion of collapsing physical servers into logical instances, it can seem, at first glance, that there is increased risk. In fact, the opposite is true - it is simply a matter of finding the right balance between cost and time to recover.
How long would it take you to recover from server failures in your environment?
Consolidating servers through virtualization has compelling economics
By leveraging more powerful processors and more memory, virtual servers can perform as fast or faster than distinct physical servers
Recovery from failure is faster in a virtualized environment
Our white paper will teach you
5 key things to consider when virtualizing!
Download the free white paper to learn how to
make painful, day long server recoveries
a thing of the past!