SQL Server Consolidation, Part 1

December 20, 2012

SQL Server Consolidation, Part 1

I am seeing more than a handful of articles these days regarding SQL Server virtualization and/or consolidation. Lots of companies are coming forward to provide you tools that will take the pain out of such efforts. They promise tools that will help manage your current capacity, help you monitor and troubleshoot virtual hosts and guests, and help you streamline overall database management.

And why, exactly are all of these companies climbing over each other to provide these tools? Because companies are looking to consolidate. Not just database servers, they are looking to consolidate as much as possible. The days of excess appear to be mostly over. It is harder to get new hardware in the door these days than ever before. If you need a new server, you had better provide a handful of reasons why you cannot use an existing piece of hardware. And stomping your feet like a five-year-old and saying “because I don’t want to share” is just not enough anymore.

You need actual reasons, not fear and speculation of the unknown. So, what are the problems driving consolidation and virtualization efforts? Why are companies looking to go through these (sometimes painful) efforts?

Cost Reductions

No doubt that this is the biggest reason cited for want to do a consolidation or virtualization project. You will see mention of a variety of expenses that will be reduced. Licensing is usually the biggest driver, followed by the need for fewer resources to monitor fewer systems. But you will also see mention of other costs, such as reduced power consumption, which ties closely to a reduction in cooling expenses. Other items that gets mentioned would fall under cost avoidance, particularly the reduction in physical servers that results in the reduced need for rack space, meaning you won’t have to knock down any walls and expand your data center anytime soon. I have seen this listed as a cost reduction, but it really is a cost avoidance. Yes, there is a difference between saving money you have been spending versus spending money that is still in your pocket. I can’t go around not buying diamonds and tell everyone I saved my household a few million dollars in expenses last year, yet I see some people trying to make similar claims. With fewer physical servers you have fewer cables (usually), fewer routers, fewer switches, etc. In other words, when you start adding in every piece of hardware and person required to maintain that hardware (accounting for their time), the costs add up quickly. It is these associated costs that become a very large factor in driving any consolidation or virtualization project.

Inefficient Use of Resources

Truthfully you could also consider this to be a reduction of expenses, but I like to think of it separately. I like to lump money into one bucket and resource efficiency into another. An example of this would be CPU usage. For example, consider a database server with 8 CPUs, but the server never uses more than two. You are essentially “wasting” six CPUs. If you were to place that database server as a guest on a host, you could start using your CPUs more efficiently. But there is more than just hardware efficiencies in play. There are also software efficiencies. With virtualization and consolidation, you start reducing the number of operating systems in your environment. As such, you have a chance to standardize your environment; you might be able to ditch those old NT servers! And you could have centralized database management. You could even start segmenting your servers in a way that you could have certain team members responsible for replication, others for the BI stack and SSRS servers, and others for performance tuning.

High Availability and Disaster Recovery

After cost reduction and efficient use of resources, there is a third common category: the use of tools focused on high availability and disaster recovery. When you begin to virtualize your servers, you start to find it easier to move things around when necessary. There are lots of tools to help you with HA efforts; I won’t mention them by name here right now, but the idea is that you can increase your up-time dramatically. In some cases, this factor alone can be enough to drive businesses to virtualize there servers. I have seen cases where virtualization is also driven by efforts to have better disaster recovery plans. Again, it is easier to move virtual servers than physical servers. Companies located in zones that have a certain frequency of natural events (hurricanes, earthquakes, Hannah Montana concerts) are always looking to ensure they have a plan in place to avert going out of business altogether.

This entry focused on the driving factors for SQL Server consolidation and/or virtualization projects. In the next entry, I will talk about the decision process for such projects.

Thomas LaRock is a Head Geek at SolarWinds and a Microsoft® Certified Master, Microsoft Data Platform MVP, VMware® vExpert, and former Microsoft Certified Trainer. He has over 20 years’ experience in the IT industry as a programmer, developer, analyst, and database administrator. LaRock has spent much of his career focused on data and database administration, which led to his being chosen as a Technical Evangelist for Confio Software in 2010, where his research and experience helped to create the initial versions of the software now known as SolarWinds® Database Performance Analyzer. LaRock has served on the board of directors for the Professional Association for SQL Server® (PASS), and is an avid blogger, author, and technical reviewer for numerous books about SQL Server management. He now focuses on working with customers to help resolve problems and answer questions regarding database performance tuning and virtualization for SQL Server, Oracle®, MySQL®, SAP®, and DB2®. He’s made it his mission to give IT and data professionals longer weekends.