Building a Better IT Architecture
I’ve spent the last 12 years working with endpoint management in one form or another. In that time my responsibilities have shifted dramatically. I first started at a technology reseller setting up workstations using Ghost. I expanded past that role into one of delivering software. First I delivered software with login scripts and batch files that were executed manually. Later, about 10 years ago, I moved to Burns & McDonnell Engineering and helped deliver software using those batch files and Netware Application Launcher (NAL). Then almost 5 years ago we switched from Novell to the Altiris Client Management Suite.
Over that time I also performed a number of other IT project related duties. Many times I was responsible for choosing a new standard application or product to meet a specific need that our organization might have. In fact I was the one that initially researched a replacement for NAL and recommended Altiris Client Management Suite.
Not to toot my own horn, but it got to the point where I could review a product and I could tell when it would or wouldn’t be a good fit for our organization. I could almost instinctively feel out if it fit into the Burns & McDonnell model of IT and how it would or wouldn’t work for us. I spent some time awhile back trying to actually quantify and categorize how I made my decisions. Surely there were invisible criteria I was unconsciously assigning to these decisions. Some of the worst products I could identify as a bad fit without even really taking time to evaluate the product in person and hands on. I’ve spoken to a number of other IT professionals, including some of the great guys at our local user group meetings, and some of the sharpest guys I know are the same way. We all sometimes instinctively know something is a good or bad fit, but we aren’t always able to say why.
I knew if I could quantify these decisions, somehow put them on paper, then I’d be able to improve upon them. Perhaps I’d be able to more quickly recognize better products. Maybe I’d see better when I may or may not be getting wrapped up in marketing madness. It would be especially important when working with technologies I wasn’t as familiar with. But alas, my process remained undefined.
Fast forward to about a year ago. For 4 years I’d been the Altiris/Symantec Administrator for our organization. I’d spent a great deal of time focusing on making the right technology decisions for our desktop section of the IT Architecture. It was shortly after returning from ManageFusion 2009 when I was offered a new position within a new IT Architecture group. Each of us had a specific focus area, mine being workstations obviously.
We spent some time trying to figure out several of the key challenges facing our organization and how would be the best way to tackle them. The director of IT, Kris Paper, sat down with us about a month in and we had a quick discussion. She shared with us how it was important for IT to become a valued business partner. It was then that she revealed to us 10 Architecture Guiding Principles designed to help us take an Enterprise view of IT and how integral it is to the business. When I read over these 10 items I realized this is almost exactly what I’ve been trying to build myself.
Below you’ll find the list of those 10 principles. I’ll start by stressing these are what we’ve defined as important to our organization. Before adopting any of them I suggest you seriously stop and read through them. Think about what is important for your organization and if they should be added to, taken way from, or even changed in some other way.
- Greater Good: Each decision should be made with the context of its overall benefit to the company and its business partners. We believe there are really only three factors in determining a business’ success: People, Processes, and Technology. If HR can bring in the right people and the production groups can develop and act on the correct processes, then the rest is up to IT. Many IT groups in many organizations don’t realize they’re a full third of the force in deciding the success or failure of any organization.
- Controlled Risk: There is inherently a risk to introducing any information technology change or service addition. It is important to carefully assess and manage that risk so as to not adversely impact the organization. Careful planning should be used to make sure the business impact is appropriate in accordance with the implementation or change.
- Support and Maintainability: Each architectural decision should be made with the future support of the newly introduced or changed service in mind. A flexible environment that can take advantage of emerging technologies is a wonderful thing however the ability to support that environment through subsequent changes and upgrades is important. It is for this reason it is also recommended to strive to minimize the number of vendors and standards in use in and environment.
- Usability: Information Technology systems in the environment should be convenient, intuitive, integrated, and minimize the need to relearn key systems so that a user is able to perform their job functions effectively.
- Flexibility and Adaptability: Incorporating business changes into existing systems is a key to overall success of the organization. It is important to value component modularity that takes advantage of standard interfaces. This allows for the ability to adjust system process while maintaining continuity for the user.
- Rate of Change: It is important to maintain the ability to quickly deploy technology to meet the changing needs of the business. Each change should be enacted with appropriate caution, but not interfere or otherwise inhibit the business from operating in any way.
- Cost Effectiveness: It is important to balance short-term and long-term costs with their corresponding benefits when selecting technology. Always strive to achieve an acceptable balance between the business benefit and the cost of implementation
- Security: Security should be approached in a flexible way that can be tailored to the specific needs of a business unit while maintaining the appropriate controls required for the situation.
- Reliability: The overall environment should remain reliable and continue to support the business processes to the utmost of levels.
- Information is a Corporate Asset: Information should be viewed as a corporate asset and used to support the business requirements of any product. Also users should be empowered to access relevant information when working within their own business processes.
Now that we’ve laid each of these out let’s take a few moments to run through an example. I’ll run through how a decision to go with Client Management Suite might look.
- Greater Good: While hard to quantify specifically it’s obvious that an Endpoint management system could be used to the advantage of an organization for all of the options it presents and information it can provide. This one passes here.
- Controlled Risk: There is obviously a risk to introducing any agent on a workstation so careful monitoring and measurement should be introduced when making a decision. In addition any regularly scheduled processes shouldn’t impact performance. Our specific environment didn’t find any slowdown introducing the Altiris Agent and the ability to adjust the schedule of inventories mitigated slowdowns there. Another pass
- Support and Maintainability: Client Management Suite is a well known product and with the array of partners and professional services available to help we can always get answers to any issues no matter what integrated module we might be using. Pass
- Usability: For users there was little to no change over what previous software delivery experiences they might have experienced. For administrators the interface is picked up with relatively little additional investigation and instruction time. Passing marks.
- Flexibility and Adaptability: Each solution uses the same interface now with increased integration on the SMP and additional products are rolling into this as time progresses. Big win here.
- Rate of Change: Client management suite is not only easy to implement and get going, but it also allows an organization to quickly and consistently reproduce reliable results when rolling out software. We’re good here.
- Cost Effectiveness: This is another hard one for me personally to balance out. I feel like we get value enough to justify it many times over. You’re specific processes for this will need to be considered. I’m good on this one.
- Security: Options exist to allow changes to access on the client and on the console. In addition there are solutions such as Local Security Solution from Arellia or SEP from Symantec that allow for better security on the endpoints. It’s good here.
- Reliability: For this we sat in on one of the KC user group meetings before we made a purchase. We spent time with other users and got some straight from the source information. It all came up positive. We also haven’t had an issue with reliability. Another good one.
- Information is a Corporate Asset: One word: Inventory. It’s been a huge help in unlocking IT value and determining costs for software products. I’m super proud of our ability to help save money on renewals with this one.
As you look through this list you may see how it can be applied to other areas as well, not just the introduction of new technology. Take some time and run it through any manual processes you might have. Is that software approval process for the "Greater Good"? Does it support your "Rate of Change" required by the business?
For fun think about that that guy or gal in the cubicle across from you. How do they rate on "Controlled Risk" or "Security"? It’s not something I’d recommend doing to their faces, but you’ll instantly recognize this can be a pretty good litmus test for most everything in IT and even business in general.