Now I’m not saying that we should never use spreadsheets at all - great things can be achieved with them very quickly, and they certainly have their place within some business functions, however as a business grows, the value they may have boils down to how much that business values accuracy, utility and rapid answers.
The spreadsheet has long been the weapon of choice for data analysts. When managing structured data in the past, the user was in control. They could quickly generate models and graphical representations of their data and share these throughout the organisation. Simply put – an effective, convenient and simple tool for what was required – but this simple tool came with serious baggage.
In the past, the spreadsheet was used by even the most advanced capacity planners. It typically involved a tedious process of pulling together data from multiple diverse sources, and it was prone to human error; the output always giving an accurate representation of the input, which unfortunately could never boast the same degree of reliability. The result would seldom be considered the single source of truth.
It would likely have taken numerous analysts weeks of effort to aggregate, unify and chart the data. As spreadsheets were personal to the individual it also gave the illusion of being a repeatable process, whilst being little more than a template, creating single points of failure within an organisation if key staff were unavailable or had moved on.
Whilst spreadsheets could deliver the information needed to form a view of current usage across an IT estate, and instil a degree of confidence that foreseeable demand could be met, decisions were based largely on the educated guesses of the capacity managers. This is simply not enough anymore - poor IT capacity planning has led to many high profile outages and IT downtime, affecting not only organisations’ customers but their reputations.
What has changed?
Increasingly complex, virtual and hybrid environments have made it very challenging for IT operations managers to effectively manage their IT estates and adequately plan for the future.
Operations’ size and scale have increased significantly and there is a distinct need for automated data collection and analysis. The number of options and interactions between the applications and services running in organisations’ environments has exponentially increased.
When capacity planning, the usual constraints of CPU, memory and storage have been made even more challenging by hybrid environments, data regulations, power users and server affinities.
Organisations still relying on spreadsheets are unfortunately leaving themselves exposed and vulnerable.
Why gamble when you have a sure bet?
The days of guesswork and ‘finger in the air’ estimations are gone – in an age when data is bountiful, decisions should be based on data-driven, quantifiable insight. A new breed of IT capacity planning tooling and a data science approach to capacity management - based on facts - is ensuring this becomes a reality.
In addition to an exponential growth in the volume of data, the past few years has seen a massive increase in its velocity and variety. These factors are also extremely important to the capacity planning process.
In terms of capacity planning, it is important to consider the velocity with which one can analyse large datasets and respond with insight. It is also essential to take account of the variety of things that must be considered, and an ability to bring all disparate, available data together in order to understand the ever-changing dynamics of the IT environment.
This is where the application of advanced IT operational analytics to IT capacity planning can help. By taking an analytical and proactive approach to IT capacity planning based on low level data captured from across physical and virtual environments, cloud-based capacity planning services can provide a unified view of your IT estate and a fresh perspective.
These services can accurately model complex, multi-faceted environments and determine the impact of growth or a technology refresh within seconds, pinpointing exactly when additional infrastructure resources will be required. They can identify complex patterns and predict behaviour, enabling proactive management and optimisation of an IT estate to support future growth and efficient capital investment.
Organisations that need to provide highly available, robust and reliable services can no longer afford to walk into the future blindfolded, and no longer need to. Start basing your capacity planning decisions on facts – kill the spreadsheet.
Peter Duffy, CTO, Sumerian