Why don't organisations consolidate their infrastructure, when they could realise large savings?
Data centres are underutilised, resulting in wasted cost
Our research confirms anecdotal evidence that datacentres are dramatically underutilised. Over 90% of the time, infrastructure is running at under 10% utilisation. This leaves the IT department struggling to justify their high, (and typically increasing), costs to the business.
The IT budget may not be under control, and you are probably finding it difficult to turn down requests for more capacity, even if they have not been fully justified.
Organisations typically don’t get the benefits that they could
One reason that organisations fail to consolidate is that they don’t understand the extent to which they are underutilised. The way that server utilisation is aggregated typically overstates the real situation by up to a factor of six. But this may be more depending on how inappropriate the measurement is in the first place.
When organisations realise that they are wasting money on too much infrastructure they are blocked by a number of obstacles. These include:
- Application owners are fearful that reducing hardware capacity will affect business outcomes.
- Potentially co-hosted applications may have different requirements, such as:
- Security requirements
- Availability requirements
- Operating systems
- Physical locations
- There is a fear that co-hosted applications will compete for resources and effect performance and therefore business outcomes
- Technologies such as virtualisation, cloud and SaaS are thought to make consolidation irrelevant
- Consolidation opportunities are often identified via CPU utilisation but memory, network or I/O may also be the constraining resource
- The complexity of the resulting co-hosting servers make management challenging
- There may be no budget to carry out the recommended reconfigurations:
- Re-coding may be required to move to a different O/S
- The ROI for carrying out the work may indicate it is not worth the investment
Even when the organisation bites the bullet, and undertakes server consolidation, they often set arbitrary and easily achievable targets and get nowhere near the maximum potential savings. It is typical to see organisations slap themselves on the back when they reduce their costs by 15%. But this ignores the fact that they could have achieved a 50% saving, or even more.
This does not go far in helping to banish the idea within the business that the IT department is a wasteful cost centre.
How could organisations take full advantage of the opportunity to consolidate?
In 2007 Computer Weekly explained that there are four methods of consolidation:
- Physical – move from distributed resources to a data centre
- Re-hosting – replacement of end of serviceable life components
- Logical – partitioning
- Workload Optimisation – automatic partitioning based on demands
Computer Weekly go on to state:
“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 analyse current opportunities and various consolidation options. Comparing the solutions' TCO and service levels head-to-head with a TCO analysis tool can provide the team with visibility into potential savings, and provide justification needed to empower the business to make the right decision.”
Capacitas are such an independent analyst and offer a service which allows organisations to maximise their savings, and even improve service quality while doing it.
Capacitas offer an infrastructure consolidation service, to maximise savings
The Capacitas infrastructure consolidation service is designed to overcome the obstacles to achieving the greatest benefits. We have repeatedly delivered cost savings of tens of millions of pounds, without service degradation.
This is done by:
- Application and service owner’s requirements should be addressed by identifying opportunities to consolidate among similar systems
- Aggregating demand by using appropriate benchmarks, (normalisation), and recommending multiple solutions based on the same benchmarking methodologies
- Considering all possible resources, not just CPU
- Applying performance modelling capabilities to assure that business outcomes can be achieved
This results in confident budget plans, being in control, (the ability to pull the correct levers), and a justification to the business of how IT can make a valuable, efficient contribution.
To learn more about how Capacitas has achieved these large savings, click here.
For more detail, please download the white paper using the link below: