Lots of IT shops are looking to grab the benefits of Cloud for the network. Jose Cornejo, Enterprise Network Architect with HP Network Consulting Services, is out there in the trenches with them every day. In this interview with Technology Services editor John Cummings, Cornejo describes how organizations can lay the groundwork for an efficient, flexible, cloud-enabled network.
John Cummings: So what does the Cloud mean for the network?
Jose Cornejo: What it means is being able to achieve a very high maturity level across your people, processes and technologies. In order to be able to gain the benefits of cloud and cloud-based services, many organizations have to update their existing infrastructure, update their policies and procedures, and train their people. It's like a three-legged stool; if you’re missing one of the legs it’s not going to work. It's not enough to have the right type of automation and business systems in place if your people and processes are not in place.
Cummings: On the people side of the equation, you’ve written that organizations that rank low on the maturity scale are often characterized by having “competency heroes.” That sounds like a good thing, but I'm guessing it is not!
Cornejo: A competency hero is the person that everybody goes to when they need to fix a problem. Whenever there’s an issue they call upon this person, who comes in and magically resolves the problem. And then everything works okay until the next time, when the hero has to come back and save the day again.
This tells you that you have some underlying issue that’s just being patched or mitigated for a short period of time. It also says that you don't have well documented processes and enough people who understand that particular architecture to be able to address the problem, because now all of a sudden your hero happens to be on vacation or decides to walk away from the job, and now the company is extremely vulnerable.
Unfortunately most organizations don't realize this until it's very late. They have a massive outage or something, and then they’re surprised to learn that this issue has been lingering for quite some time. When the rest of the staff are asked to step in and help, their response is: “What? We don't know anything about it because this other person always did it.”
Cummings: Once a company has moved beyond that stage and it has the right policies and people in place, what's the next stage as it continues to develop its maturity?
Cornejo: Once the documentation is in place it has to be validated. It's not enough to have documentation if the information is either outdated or inconsistent. You have to validate that the information that you have outlined is actually applicable to the situation. It has to be tested with someone who did not write the document, someone who’s looking at it for the first time and is able to make sense of the material and apply it to a practical problem.
Cummings: There has to be an ongoing check on the information, some kind of institutionalized way of making sure that it sticks.
Cornejo: Absolutely. The other thing that you want to do is to make sure that it's in a public place where it's easily accessible. One of the things I run into is that there’s great documentation, but when you ask “Where is the central repository?” the answer is “Well, there is no central repository, it resides on the hard drive of an individual who is not in the office today.”
Cummings: So you’re basically back to square one!
Cornejo: Exactly. Unfortunately most organizations don't have a centralized repository, for a number of reasons, and that’s one of the biggest issues that can actually bring an organization down. It also prevents them from being able to build beyond what they've already have. If an organization in the most fundamental sense lacks documentation to represent what they have in place, what leads us to believe that they will have any collateral to share or to be able to review as they work at making projections for the future? How can they create a blueprint for what they want to do moving forward? This is one of the things that lead you to suspect that, from a maturity perspective, they are still in the early stages.
Cummings: What does the organization look like as it approaches maturity and readiness for cloud services?
Cornejo: Well, you think about a utility-based model where you have appropriate instrumentation in place, you have backups and recovery mechanisms in place, you have people who are adequately trained and skilled and who are able to address issues from a best-practice perspective. You have well documented and up-to-speed documentation information that’s held in a central repository and backed up on a regular basis so that if that particular media becomes unavailable you can quickly and easily restore it.
It also implies that you have processes in place that you follow and that you can execute upon in the event of an emergency across multiple areas. It implies that your system has alternate or recovery paths that are automatic so that if an aspect or a portion of the network goes down, the system will automatically recognize that there’s a failure and reroute traffic via an alternate path without any manual intervention.
Cummings: Another characteristic of this state is that it’s “business-centric.” Can you expand on that a little?
Cornejo: This means that you have a good understanding of the key business systems and the cause-and-effect relationship with your network. For example, if you rely on several high-speed connections to some key mission-critical systems, and half of them go down because of a device becoming unavailable, you understand the impact to your business systems. Is that going to drive utilization up 100 percent? Is that proportional? Are you running at 25 percent or 50 percent on both sides, so that should you break one link you can run at the same level? Or will you be running at a reduced level, where you’re able to process key business applications, but other things that are considered less mission-critical are going to be slowed down?
Cummings: You know how soon you would reach a point where you’re going to start losing money.
Cornejo: Yes. These are the things you need to understand. It’s really important that the IT organization should develop its business acumen and the ability to offer this kind of insight to the business at large, but it generally takes time to achieve this. Maturity doesn’t arrive overnight. IT teams that decide to accelerate this aspect of their development have a tremendous opportunity to add value by building a solid understanding of how the decisions they make relate to the business systems that they’re making available.
Once you reach that point, you’re able to have a businesslike discussion with the leadership team to help them understand that failure to invest in certain areas will result in x amount of revenue loss because all of a sudden your transactions are running at half speed. You’re able to produce adequate business cases to explain why an investment needs to be made.
We’ve found that the most successful IT organizations understand that they have a vested interest in learning more about the business so that they can have a better appreciation not only of how the company works, but of how IT works for the company.
That level of business acumen gives you the opportunity to better serve your customers, which is part of the reason why most executives who move up in the IT area are folks who come up through the application space. They learn about the business because the applications essentially ride whatever business processes they’re needed for, so it’s no surprise they can more easily move into key roles. But there’s no reason why the strictly technical people, the folks who deal only with the infrastructure, shouldn’t have the same kind of career opportunities.
