In the good old days, IT operations managers were responsible for maintaining the infrastructure, meeting service levels agreements, sticking to budget, and keeping employees happy. Life was not easy, but at least it was familiar. You knew your hardware, your software, your employees. You determined services levels based on what you could actually see and touch. You told people what to do and they did it. While IT was perceived to be an expensive cost center, it wasn’t an issue as long as the phones worked and the systems ran according to expectation.
That was then, and this is now. Today IT operations managers can still see a lot, but they can touch very little. While some companies still own a significant amount of hardware, a diminishing number are making new hardware purchases. They’re moving toward Infrastructure as a Service (IaaS) solutions. IaaS makes hardware, storage, network, software, and all the work required to support such things opaque to the consumer, very much in the same way that the power generators at Niagara Falls become opaque to those using electricity. All the consumer sees are the power outlets in the wall and the monthly bill sent by the power company. By the same token, IT operations managers only see control panels and dashboards. The days of going over to the data center to inspect capital assets are gone. Now, procurement is about getting the best hourly rate for a given service. It’s a different way of doing business.
Not only is the nature of hardware in the enterprise changing—software is too. Given the proliferation of using continuous integration and continuous delivery techniques to deploy a greater number of smaller, container-centric applications and services, today’s IT operations require less human interaction to make software available to end users. Most of the work is done by scripts and in some cases, scripts made by other scripts.
Applications are becoming more fine-grain. The monolithic application is giving way to those that are composed of collections of microservices deployed as ephemeral containers configured and controlled via orchestration. Increasing the number of fine-grain deployment units and the rate at which they change decreases the likelihood that one person will know every detail of an application. As applications and services become smaller, the technical authority and responsibility associated with them becomes more decentralized. Whereas in the past a “department” owned an application, in today’s world of containers and microservices, small, autonomous teams are the responsible entities. The role of the IT operations manager is no longer that of the expert gatekeeper, but rather that of the knowledgeable facilitator.
So then the question becomes, what are the essential responsibilities of the manager in modern IT operations? There are 3 essential responsibilities:
- To set realistic budgetary expectations
- To ensure and maintain IT operations transparency
- To mitigate the impact of automation on the workforce
Allow me to elaborate.
Set realistic budgetary expectations
One of the benefits of implementing cloud-based IaaS and container-based architectures deployed by way of automation is considerable cost saving. According to David Bray, CIO at the Federal Communication Commission:
“Back in late 2013, we were spending 85 percent of our budget just to maintain 10-year old legacy systems. However, after a dramatic shift to public cloud and commercial service provider, we saw our maintenance spend drop down to less than 50 percent of our budget.”
That’s right, 50% savings. This is no small amount. It becomes the stuff of legends. It also becomes the standard by which expectations will be set if left unchecked. It’s only natural for stakeholders to think, “We had 50% savings this year. Getting another 50% savings next year is a fair expectation.”
Yes, automation will reduce costs dramatically at first, but then over time the saving will flatten out. Big efficiencies and cost savings are typically achieved early on. It’s a typical pattern when moving to cloud-based infrastructures. Yet, many in upper management are accustomed to making an up-front investment in an initiative with the goal of long-term savings. Thus, those in upper management who are new to cloud-based transformations might treat the significant saving incurred initially as an indicator of good things to come, that greater saving are on the way. Yet such an expectation is inferred and unrealistic. Thus, the wise IT operations manager will do well to keep budgetary expectation fact based. The wise IT operations manager will ensure that the immediate and projected costs and benefits of his or her organization's activities are clearly stated and available to appropriate parties easily, on demand. In other words, transparency is everything.
So then, how is transparency achieved? Read on.
Ensure and maintain IT operations transparency
One of the benefits of IaaS is the rise in reporting technology. Gone are the days when it took a day’s worth of labor at the end of the month to create and distribute budgetary information through the management chain. In fact, IaaS companies such as AWS, Microsoft Azure and Google Compute Engine make the latest usage and billing information available to consumers at the click of a button. Today stakeholders can be in the know all the time and should be. As a result, stakeholders throughout the enterprise want to be in charge of their information infrastructure. To quote Laurence Chertoff independent consultant to nonprofits and former Director of IT at Npower Inc.:
“If the end user cannot modify or administer the technology quickly, I don't want it.”
It doesn’t matter whether a technology is out on AWS or in a server room on the 3rd floor, people want access to their technology. Thus, instead of being a powerbroker of information dissemination, the role of the modern IT operations manager is to make sure that people can get at the information they need, when they need it, in a manner that is easy, appropriate and secure. In addition providing the transparency that allows users to do as much for themselves as they can and want to do, items such as incident tracking, operational costs, service issues, and project workflow data need to be apparent and accurate. At the end of the day, the modern IT operations manager is a facilitator, not a roadblock, to information transparency. Making systems flow with as little effort as possible is one of the the essential responsibilities of modern IT operations management.
Mitigate the impact of automation on the workforce
Automation has an impact. Always has, always will, ever since the days when Dutch windmills ground wheat into flour faster than a human ever could. The good news is that automation technology made it possible for many more people to have more bread at a cheaper price. The bad news is that people who manually ground wheat into flour had to find other types of work to do in order to survive.
The same holds true in IT operations. For better or worse, increasing the use of automation in IT operations means that the way people work has changed and will continue to do so. Steve Mays, CTO of Trizic has an interesting take on the matter:
“Where we used to depend a LOT on the ‘art’ of super talented people who had a widely varied background from operations to development to help us build and manage systems and software, we now have automation in the cloud. In the past, [engineers] hand crafted the product to be artisanal and got a lot of satisfaction from a job finely crafted. [Today] we look at IT operations more like an assembly line. My mission is to turn out 100K units vs. beautifully [releasing] only 10.
From a product output and quality perspective, this is a good thing. However, the systems are no longer your ‘friends’ that you have a ‘close personal relationship to’.”
IT operations will become more automated, not less. The use of human labor will fall into two categories. One category is that work which requires creativity and imagination, for example system design and incident troubleshooting. The other category is work which is predictable and repetitive—requirements gathering and capacity planning. Eventually all predictable and repetitive work will be automated. It’s the nature of the dynamic. If machine intelligence can observe a pattern for a long enough period of time, eventually the machine intelligence can emulate the pattern—think speak recognition.
Thus, the modern IT operations manager will be subject to a pressure that is new to the IT operations—how to plan for the continuing obsolescence of a portion of the workforce. To date, IT operations managers worrying about employee obsolescence has been minimal. The conventional thinking is that IT operations employees have always been in demand. If a job is lost in one company, a new job can be found in another. But, as comprehensive automation becomes more pervasive, the notion that there will always be another job might be a faulty assumption.
Of course, the IT operations manager can simply not care. The problem is you can’t fake sincerity for a long period of time. As more automation sets in and employees face increasing pressures to exert more creativity or work with new, more difficult technologies in order to stay viable, many will turn to management for help. If that help appears as a lot of talk with no action, eventually employees will figure it out. The result is a workplace in which morale low and personal investment is nonexistent. In such an environment the quality of service degrades and eventually costs increase. Not caring is a solution with a very shallow horizon and poor ROI.
The alternative is to take a new approach. Modern IT operations management is about taking the time to understand the impact of automation upon the workforce and having the foresight to take realistic steps to address the impacts early on. At the least, the impact of a given automation event needs to investigated and articulated in a transparent manner, particularly with regard to the human impact. Then, realistic strategies to address the expected impact need to be devised. If such strategies require accepting the substitution of human labor with automation, even if the substitution means the cost of an employee’s job, the enlightened IT operations manager will promote such discussions and foster ways to find solutions for the problems at hand.
In the old days, the IT operations manager would just pass the issue of impact of automation on the IT operations workforce onto HR. However, given the decentralization of both authority and expertise brought about by automation itself, HR is no longer the expert in these matters. HR simply does not know enough. The IT operations manager is now the authority and expert, even in matters of personnel.
Mitigating the impact of automation on human employment will be a growing concern of managers in IT operations. The issue is new and daunting. Yet, for some managers it will be an exciting opportunity. Those that can succeed in finding ways to mitigate the impact of automation on those they manager will stand at the forefront of the profession.
Putting It All Together
In the world of IT operations, the old days of manage by edict and command are over. Steve Mays from Trizic sums it up well:
“Top down is 100% impossible now. My 30+ year background in tech isn’t that useful anymore. I used to guide folks with my experience. Now I guide them to exercise restraint, build good enough features but with high levels of testing to ensure fewer issues. I let them tell me all about the new frameworks/libraries/tech that they think we should use.”
Today, given the continuous demands that enterprises make upon IT operations, IT operations managers need to provide to their stakeholders more complex services, at faster rates of implementation and higher levels of reliability. Automation technologies combined with greater transparency into IT operations are key elements for meeting these demands. Also, given the growing trend to decentralization, IT managers can no longer be expected to be know-it-alls. Real knowledge and authority resides in the small teams providing the solutions needed. The job of the modern IT operations manager is to be a knowledgeable facilitator, one who establishes and maintains an appropriately open and transparent environment from which realistic performance and budgetary expectations can be established. The modern IT operations manager sets the guidelines in which good work can happen. Also, given the growing prevalence of advanced automation technologies on the IT landscape and the impact that such automation will have on human employment, the modern IT operations manager will at the least, articulate the human ramifications of automation initiatives at hand and foster an environment in which reasonable discussion about the impact of automation can take place.
The modern IT operations manager understands the dynamics of technical authority within the decentralized enterprise and the importance of providing timely, accurate information to support such authority. He or she knows how to both inspire and guide others to find the tools and techniques necessary to allow the enterprise to compete successfully in the marketplace for the benefit of all it touches—customers, employees, and shareholders.
The modern IT operations manager understands that the real power of his or her position is to empower others to act reasonably, affordably, and safely. This is no longer power expressed by edict. Rather it is power that comes from influence. Power expressed as edict has the shelf life of a manager’s tenure. Power that comes from influence lives on well after the manager has left the company. I’ll leave it to the reader to decide which type of manager he or she wishes to be.
For information on Rapid7’s IT operations solutions, click here.