Most organizations today have far more servers than they actually need. Low buy-in costs have made acquiring more servers a cheap, easy way to provide local connectivity, solve capacity problems, and meet immediate needs. After all, how much time can an organization spend analyzing a low-cost purchase? The result has been a proliferation of devices, not only representing a significant investment by themselves but also requiring software, power, connections, and management time. Large numbers of servers must be maintained, their software licensed, configured, and updated, and their security ensured.

In practice, according to Forrester Research, Windows servers typically run at only 8-12% of their full capacity, and UNIX servers at only 25-30%. Clearly, this represents a tremendous consolidation opportunity and a chance to save large amounts on server purchases, energy, administrative costs, space, cabling, licensing, and accessories. Server consolidation is thus an easy sell in almost every data center. License redundancy alone can often save enough money right away to recoup the costs. A typical survey recently reported an average savings of 16% from server consolidation projects. However, actually identifying underutilized servers and combining their operations is not as easy as one might think. Where do you start and how do you proceed?

For example, you could use network discovery, asset management, performance monitoring, or other widely available tools. But applying them to large data centers with thousands of servers typically involves a tremendous amount of administrative time and manual analysis. Automation is therefore essential to pinpoint consolidation targets. Furthermore, one-time analysis won't do the job. You must monitor continuously, looking for opportunities all the time.

Effective server consolidation requires extensive knowledge of both static and dynamic attributes. You must understand the target systems in detail, including hardware inventory and configuration, operating system settings and files, installed patches/hot fixes, application inventory and configuration, middleware configuration, database configuration, system capacity and utilization. And this is just the beginning, because comprehensive analysis necessitates scrutinizing system configurations and runtime behavior more deeply.

There are three key steps to identifying consolidation opportunities:
1) Understand what you have and how it is configured.
2) Perform detailed configuration comparisons across systems to identify synergies and similarities
3) Obtain detailed workload information to enable "what if" analyses

Tools are available to help with the process. For example, CiRBA's Data Center Intelligence product combines discovery, configuration auditing, workload profiling, and specialized consolidation reporting, all in a non-intrusive form. CiRBA's analysis capabilities use both deep configuration audits of hardware, software and database settings, and detailed workload profiles to identify the best candidates for consolidation. .

A strategy based on continuous data monitoring and consolidation analysis increases the likelihood of achieving substantial ROI from server consolidation. Automated tools are essential for large data centers. A quantitative, statistical approach delivers empirical and accurate results, critical to the planning and execution of successful projects.