By Ephraim Baron
I’ve spent a large part of my career in information technology, mostly focused on IT operations. I tell people it’s at least as hard to make something run reliably as it is to get it working in the first place. (Many developers disagree with me, but they’re the ones who’ve never run production operations. Perhaps the emergence of DevOps will help bridge the gap.) I also say that if you don’t have a thick skin, you shouldn’t be in Ops. That’s because when things break it’s your fault. When they work, it’s because you didn’t screw them up.
Put another way, a semi-adversarial relationship often exists between development (Apps) and operations (Ops). The main reason for this is that Ops and Apps are measured and incented differently.
Ops is supposed to provision, operate, and maintain IT infrastructure with the highest availability at the lowest cost. In short, their primary focus is ‘cheaper’.
Apps, on the other hand, is charged with delivering capabilities to the business as quickly and efficiently as possible. Their focus is ‘faster’.
The result is a fundamental disconnect. So even though both groups may be doing their jobs as expected – and even at a high level – the company may not be getting the full value it expects from IT.
The Person Who Says ‘No’
Because of the way my charter, funding, and success criteria were defined when I was running Ops, I often had to deny requests from the business. Yet despite what my customers may have thought, I got no joy from telling them ‘no’. I usually had no choice. Here are some typical examples:
Q: “Do you have a spare server I can borrow?” [… starting this afternoon. Oh, and it needs to be a really big one.]
A: No
Q: “Can I get a system for just a couple months?” [… or for as long as I need it, but absolutely no longer.]
A: No
Q: “Is there a cheaper option for storage?” [… or can I just go to my local electronics store and buy something, where I’ll be storing mission-critical data?]
A: No
To IT operations folks, these questions are as ridiculous as they are familiar. Yet to developers they seem perfectly reasonable.
Shifting to a New Consumption Model
How can we solve this seemingly intractable problem? We can’t expect the business to move more slowly or to develop more accurate forecasts. Nor should Ops anticipate a substantial increase in budget to fund “business agility.” If internal constraints are going to remain the same, the answer needs to come from somewhere else. Enter, cloud computing.
The National Institute of Standards and Technology (NIST) offers the following definition of cloud computing:
Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources that can be rapidly provisioned and released with minimal management effort or service provider interaction. It has the following essential characteristics:
- On-demand self-service
- Broad network access
- Resource pooling
- Rapid elasticity
- Measured service
If we apply these characteristics to the questions above, it looks like we just might have a solution to our IT challenge.
I frequently draw a comparison between information technology and electricity. Back when electricity was still relatively new, consumers were forced to know the details of how it was generated, transmitted, and delivered. Eventually standards emerged, central generating facilities were built, and the consumption of electricity shifted to a utility model. These days we don’t concern ourselves with how electricity reaches us. We just want a reliable supply at a reasonable cost. After all, it’s not the electricity itself that’s interesting – it’s what we do with it that matters.
Think of how inefficient it would be if every consumer of electricity were forced to design, build, and operate their own power plants. And if that power plant were to go down, the business would cease to function. Yet that’s precisely the situation we currently find ourselves in with IT. Nearly every company has their own staff and runs their own systems. And if those systems go down, the business ceases to function.
Embrace the Force, Be the Hero
So if the cloud is truly a better way of delivering IT services, why is it slow to catch on? Simply put, between people, process, and technology, the technology is the easy part. Getting people to share or even relinquish the knowledge they’ve painstakingly acquired and that they jealously guard is not trivial. Additionally, new systems require new skills. Similarly, adapting long-established processes and governance requirements to new delivery models takes time.
Yet in almost all things – and particularly in the world of IT – change is inevitable. And, as Equinix’s CEO Steve Smith is fond of saying, “cloud is irreversible”. Because after all, if given a choice between on-demand, elastic, pay-as-you-go services and slow-to-provision, hard-to-explain, hard-to-measure services, who would choose the latter?
So if your job involves IT operations, the best thing you can do – for both yourself and for your company – is to welcome cloud computing with open arms. Learn to talk in terms of services and business value rather than techno-minutiae. Stop telling your customers how difficult technology is. They don’t care. What they do care about is helping the business function and grow. Move from guardian of the ivory tower to enabler of the business. By doing so, you’ll be seen an essential contributor rather than just overhead.
In short, you’ll be the hero.
Fail to change, and you’ll be irrelevant.
Subscribe to the Equinix Blog