For the last couple years, as I have worked and focused on capturing and communicating the value of server refresh, I have been asked about the sensitivity of the analysis for different degrees of consolidation. For me the tradeoffs of how you consolidate and the relationship to the business benefits have been intuitive and I could easily explain these concepts to people I talked with. BUT, as they say, a picture is worth a thousand words.
It is obvious that the more you consolidate, the more you save and the faster ROI benefit - equally obvious was the fact that if you replace servers 1:1, there is not a TCO benefit but there is a maximum performance gain. The challenge in the explanation was what about if you did only a little consolidation and that changes the benefit equation.
Personally, I have struggled with capturing this relationship visually. However, a colleague of mine and I were talking today and he captured this well bringing me one of those "Ah Ha" moments that I just had to share.
Let me explain this chart a little. Each scenario shows the relative performance improvment and relative operational cost savings (TCO) of consolidating older single-core Intel Xeon 3.8 GHz servers purchased in 2005 or 2006 with a new Intel Xeon processor 5600 series server released in March 2010. The calculations were made using the Intel Xeon Server Refresh Savings Estimator that was developed with key learnings from the Intel IT department.
Because every situation is different and because your cost assumptions will be different, I invite you to customize your own server refresh scenario and discover the benefits. You can change over cost 50 variables, refresh with or without virtualization and change the begining and end server configuration in the tool to get the answer that is right for your organization.
I hope that what you find helps you realize the same thing that Intel IT finance team did last year ... that "defering our server refresh story was someting we could not afford to do"