Intel estimates1 that one-third of the servers in production are more than four years old. At first, one might think that it is great to get this much service out of a capital investment, but the operational costs to run these outdated servers would pay for a complete technology refresh increasing performance and reliability while reducing total costs. How is this possible? With the Intel® Xeon® Processor E5-2600 v2 product family and Cisco’s Unified Computing System.
Even servers circa 2009 can be replaced to your benefit. Refreshing from Intel Xeon Processor 5500 series servers to E5-2600 v2 based servers produces a consolidation ratio of ~4:12. But all new architectures are not equal! Cisco demonstrated this with an industry leading seven world-record benchmarks with the E5-2600 v2 processors. I compared the TCO of consolidating 64 Xeon 5500 based HP ProLiant BL460c G6 servers to a new HP ProLiant BL460c Gen8 solution using Virtual Connect FlexFabric and HP ToR 10Gb Ethernet and 8Gb Fibre Channel switching or a Cisco UCS B200 M3 solution with SingleConnect. While both the HP and Cisco solution proved to be more economical than continuing the legacy environment, the Cisco solution had a payback period of only seven months compared to 18 months for HP. A Cisco UCS solution could save you over $200,000 more than continuing with the legacy HP G6 servers over three-years.
The slide above summarizes the analysis conducted and the presentation below highlights additional business benefits of UCS. The legacy HP ProLiant BL460c G6 environment contains only OpEx costs and all 64 servers are replaced in year one using the 4:1 consolidation ratio.
NOTE: These results do not include the benefits of the UCS Advantage Trade-In Program where Cisco is offering customers up to a $650 credit for HP G6/G7 servers and up to a $1,600 for HP BladeSystem c3000/c7000 enclosures. Contact your authorized Cisco Partner for complete program details.
Want to learn more about the Cisco UCS differences?
- Cisco Unified Computing System (UCS): Changing the Economics of the Datacenter
- Cisco UCS Advantage
- Cisco UCS HP and IBM A Blade Architecture Comparison
- The One and Only Cisco UCS Manager
- Cisco UCS Blades Deploy 77% Faster Than HP Blades
- Cisco UCS Migrates Server Identities Faster and Easier than HP
- Cisco UCS Outperforms HP and IBM Blade Servers on East-West Latency
- Cisco UCS vs. HP BladeSystem Power Efficiency Comparison
- Source: Intel Planning Guide, Modernizing the Midsize Business IT Infrastructure http://www.techrepublic.com/resource-library/whitepapers/planning-guide-modernizing-the-midsize-business-it-infrastructure/
- Cisco analysis of average SPECint_rate-base2006 performance improvement between the Intel Xeon E5-2697 v2 vs. Intel Xeon X5570 rounded up from 3.7 to incorporate other system and software improvements beyond those measured by the benchmark.
Bill, when you are going to slam your competition and spew numbers on an open Blog, its considered polite to explain how you came up with those numbers.
Your chart shows HP Costs more in both CapEx and OpEx yet you have ZERO collateral to back this claim up. Care to share the numbers?
Dan,
Thanks for taking the time read the blog and thanks for identifying yourself as being with HP.
If you look at the attached SlideShare presentation, it has all the data I used to build the model. I’m sorry I didn’t make that more clear in the post. Let me refer you to slides 24 – 26. If you will use the full screen mode, the numbers on slide 26 will be clear vs. the eye chart in the embedded version.
Bill