Does Capacity Planning Matter Anymore?
Some sages, particularly in the distributed systems space, like to say that capacity planning isn’t necessary anymore. Hardware is cheaper and virtualization makes better use of resources. Besides, no one seems to know how to do it these days.
But the sages are wrong. In many ways, capacity planning has become much more essential, but the role has expanded to cover capacity in more areas and to include cost analysis. It’s harder, but also much more important.
Where once people worked on terminals attached to a mainframe, now users come in from anywhere, anytime and may hit a variety of servers in executing work. The complexity is daunting.
The Evolution of Capacity Planning
We used to have business partner predictions on capacity. While they weren’t perfect, you could rely on the optimists to predict a sales campaign would go really well and pessimists to figure no one would care about the new application. You could manage that.
Distributed systems arrived and applications started mining mainframe data and later applications. No one on that side of the house ever warned the capacity planner that the volume was going to hit.
You had to be watching, asking questions, evaluating data. As your customers moved from web to mobile-enabled devices and apps grew to put more demands on your back-end systems, the challenge grew. These users expected great performance and weren’t terribly loyal. If you couldn’t respond, they’d go somewhere else.
Still, the hardware vendors kept telling you hardware is getting cheaper. You don’t need to plan. And yet, if you didn’t plan, the impact could be severe. Companies were loath to spend money on more hardware unless you proved the need, but it was harder to find the data to do the job.
With hardware vendors suggesting that capacity planning was dead, many people found they were asked to take on other kinds of work. “You don’t need to spend all your time doing that job anymore,” they were told.
And software prices kept going up and became a bigger concern. Capacity planners had to consider software placement and utilization, even with VWLC, to keep the costs from growing. In many shops, the job has grown to also look at space, cooling and electrical capacity planning.
While mainframes were never a huge electrical draw, the server farms have pushed some data centers to disperse hardware. Some cities are out of electrical capacity (London, New York and more).
Looking to the Future
The “Internet of things” (IoT) is going to drive more capacity issues as appliances in people’s houses start making demands on the system. You won’t know the impact in advance. It will just hit your system, potentially impacting other critical applications.
The solution? Document the issues and impacts for your management team. In doing so, you may not only justify your position. You’ll also be able to demand software to assist you in this complex process and the training to design and perform the new job of capacity planning.
Capacity planning isn’t dead. With the right approach, the phoenix will rise from the ashes evolving into a larger and more important discipline we can all enjoy.
Latest posts by Denise Kalm (see all)
- Breaking Down Dev and Ops Silos with Communication, Collaboration and Trust - April 19, 2018
- What Do Performance and Capacity Roles Have to Do with DevOps? - February 1, 2018
- How Developers Can Boost DevOps Through Automated Batch Processing - December 7, 2017