Since the dawn of the information age, IT has been an outward looking function; using technology to support the business and allow it to achieve its objectives more efficiently and effectively.
From automating key business processes to allowing near-instant communication with customers, partners and co-workers worldwide to using virtualisation, and later the cloud, to provide a more flexible, agile approach to business.
One of the greatest changes in recent years has been the rise of Big Data; uncovering trends and informing actions in a way that was never possible before, except through extremely lucky guesswork.
From predicting market movements, to understanding precisely how customers travel around a store, to identifying the best areas for future business expansion, the ability to enable data driven decision making has made the IT department more valuable than ever. However, could the IT department be using its Big Data abilities closer to home?
Looking inwards:
For example, take the data centre; traditionally a substantial cost centre and the cornerstone of any IT strategy. Whether a large enterprise managing its own data centre or a colocation provider offering space to a variety of customers or a cloud provider using its data centre to house its services, understanding the potential impact of every change made to the data centre could be invaluable to the business.
For instance, what technology will provide the best performance at the best cost? Will the data centre still provide value for money at 90%, 70% or even 50% capacity? What investment will the data centre require over its lifetime? Essentially, with the right data, organisations could make sure that their data centres provide the best value to the business at the lowest cost.
The devil is in the data:
This doesn’t mean that an IT department will automatically be able to aim its Big Data capabilities inwards and provide actionable intelligence for the business.
Content from our partners
As with any other field, data itself is just a raw material. It’s the way in which you analyse and select that data which makes all the difference. Without this understanding, data can be not just uninformative but actively misleading.
To return to the data centre example, the average data centre produces hundreds of items of data every second, from multiple sources. Trying to collate and analyse these items to make informed predictions about data centre behaviour and costs would take days, if not longer, meaning any prediction made from them will be either significantly out of date, or rely on calculations that have been simplified to the point of untrustworthiness.
After all, how do you know which of these hundreds of items of data you can safely discard without affecting the final calculation?
At the same time, the tools used to analyse this data have lagged behind the rest of the business, with the humble spreadsheet being, until recently, the preferred option. Data Centre Infrastructure Management (DCIM) tools are now being pressed into service as the next step, yet despite their improvements even these are usually not designed with data driven decision making in mind, and so act as a stop-gap measure rather than the ultimate goal for IT departments.
Evolution
Just as retailers carefully analyse the most relevant data from their stores and customers and turn those into immediate insights using highly specialised tools, so we are seeing the evolution of tools and techniques that will allow IT departments to predict the performance and cost of their data centres with the same immediacy. As these tools enter greater use, we are likely to see a corresponding change in the ways that IT departments measure their internal operations.
For instance, one of the rules of thumb for measuring data centre performance has traditionally been Power Usage Effectiveness (PUE); or how much of the energy a data centre consumes is translated into actual computing output.
However, while useful for designing a highly efficient data centre, it doesn’t answer many of an IT department’s more pressing questions. An example would be – what value does the data centre produce? What does and will it cost to run? And just what is the cost of providing a single service, such as email?
IT has a long history of championing and supporting the rest of the business in data driven decision-making. If it can turn its expertise inwards and understand which data is relevant, how to gather and analyse it, and what answers will be the most useful, then it can ensure that it provides not just statistics but actionable intelligence.
