A typical data center spends more than 70% of the total cost of ownership (TCO) for labor or outsourced services. High costs, especially when they account for administrative and support tasks, are a key reason that few IT dollars remain for innovative projects that can help deliver competitive value.
Server consolidation is one of the most effective ways to lower the TCO of a company's data center. Here are four strategies for consolidating server assets. They can be applied independently or simultaneously, based on the following scenarios:
Physical consolidation: Collecting servers distributed across multiple remote/branch offices and business units into a central data center.
Re-hosting: Porting from older legacy platforms and operating systems to newer systems, as fewer new high-performance systems are typically needed to support the workload.
Benefits: Savings include the elimination of older -- and often expensive -- support and maintenance contracts. And since the cost of administrators and support labor is correlated to the number of servers, fewer servers means less overhead.
Additional business value comes from enhanced availability, security, management features, performance and upgrade options -- improving adaptability of the server infrastructure.
Risks: The application may require porting to another platform, depending on compatibility with prior systems. Resulting costs can come in the forms of custom code rewrites, procedures and data migration. Porting costs are often underestimated and can be expensive.
Logical consolidation: Establishing hard partitions for the operating system, application, processors and memory requirements on a single server or pool of clustered servers, so individual server "islands" are pooled onto a single server or cluster.
Benefits: Headroom is reduced by 40% or more, and companies save administration costs proportional to the lower server assets. As needs change, the team can change the partitions to allocate more resources as needed for workload demand changes and upgrade a single cluster to support overall workload demands, instead of physically managing moves, adds and changes on multiple "islands" of individual servers.
Risks: Manually managing the hard partitions is difficult in a dynamic environment where workloads change frequently, potentially introducing management burden and complexity. Many business units will not support logical consolidation where they must share servers and will need to be "sold" on its business merits and service levels. Having more eggs in one basket increases the importance on availability and business resilience.
Workload optimization: Configuring the server operating system or third-party utilities to intelligently manage server resource allocation based on workload demands. Partitions can be established based on demand and schedule rules, and the system automatically meets needs.
Benefits: Fewer CPUs and fewer servers to support multi-application workloads. For multi-application portfolios, this approach maximizes asset use and consolidation, reducing software licensing requirements, facilities costs and labor. This saves 40% or more based on the application profiles. Typically, fewer smaller applications with peaks at different times drive the highest consolidation. Because the system manages workloads and partitions, administration and support are minimized.
Risks: Establishing workload optimization configuration and rules will take some time and can be complex, requiring the help of professional services. As with logical consolidation, business unit apprehensions and business resilience best practices apply.
Benefits: Bottom-line labor savings can reach 10% by reducing complexity and standardizing purchases, configurations and management best practices. Other key benefits include improved configuration control by restricting server access and strengthened business resilience.
Costs to implement: Low. They consist of network enhancements to support the centralization, data center build-out to support the consolidation and physically moving the servers.
Risks: Performance degradations because of poor network planning and business resilience by having all of the server assets in one basket, particularly if the data center does not have adequate recovery plans.
|App. portfolio scenarios||Apps. in the portfolio||CPUs needed during normal ops||CPUs needed during peak ops||% of apps that peak at same time of day||CPUs w/o workload optimization (Islands scenario)||CPUs w/ workload optimization||CPUs saved||% CPUs saved|
|Few small apps||5||.05||2||50%||10||7||3||30%|
|Many medium apps||10||2||8||50%||80||50||30||38%|
|Few apps where two workloads peak at once||3||1||5||66%||15||11||4||27%|
|Many small apps||10||.25||.75||40%||8||5||3||38%|
|Few large apps||2||5||2.5||50%||50||30||20||40%|
|Many apps, workloads peak at different times||10||2||6||25%||60||30||30||50%|
Using workload analysis, standard and workload optimized environments show where the maximum consolidation savings could be achieved via workload optimization.
The ROI analysis
The right consolidation decision takes careful analysis of current TCO, proposed consolidation options and architectures, required investments and potential savings. Because the analysis is complex, internal IT teams should consult with independent analysts and performance benchmarking sites (such as www.spec.org) and put vendors to task (with requisite scrutiny), to help propose and analyze opportunities and various consolidation options.
Comparing the TCO and service levels head-to-head with a TCO analysis tool can provide the team with an idea of potential savings and justification needed to make the right business decision.
Tom Pisello is the founder and CEO of Orlando-based Alinean, an ROI consultancy and software provider. He can be reached at firstname.lastname@example.org. For more information and exclusive access to server TCO analysis tools and white papers, visit www.alinean.com.