Three Design Goals for Tech Infrastructure

I seem to have this conversation many times, so I thought it would be useful to write it down. Ultimately I’m a very lazy person, and will work very hard to avoid work. This is part of that plan…..don’t need to repeat the conversation, just look at my blog!

I have been a network infrastructure architect in various technical areas in a large enterprise for many years, and I’ve learned there are three key goals when designing a new system. These goals may appear to be more personal in nature, and there’s some truth to that. One interpretation of Work/Life balance would be “self preservation.” If I am happy, healthy, and have some kind of life outside work, I will work harder, enjoy life, and live longer (ultimately allowing me to work more years). Having said that, these personal goals are consistent with design goals that ultimately provide value to the business.

The first goal: Don’t get called at 3 AM, ever.

What does this mean? Design good infrastructure. Pick good products, and test them well in the lab. Put together good design docs and job aids, and train all the stakeholder teams on the technology. Follow KISS (Keep It Simple, Stupid); complexity can be your enemy. If requirements dictate it, put HA and DR in your designs. Make it good, simple, robust. You do all of this, and support teams will not need to wake you up in the middle of your night to request help.

The second goal: Remember you may be handing this off to a green employee right out of school/training. ​

What does this mean? As an architect, you are charged with being the technical expert (the Subject Matter Expert, or SME) on the technology you design. However, at the end of a project, you must hand this off to an operations team who might not know a lot about this technology, and who is doing support for many infrastructure components (not just this new one). While many support teams are great folks and very sharp, you cannot expect everyone to be a rocket scientist on your technology. And, after the project is over, you have to move along to the next technology project. I cannot put enough emphasis on KISS here. Often architects will devise very creative, unique solutions that on the surface appear to save us money by reducing capital spend or the like. In reality, complexity injects its own costs, and these are much harder to track. Complex infrastructure is harder to troubleshoot and often take more SME time to troubleshoot.

The third goal: Don’t break the bank.

What does this mean? If you do the architecture job correctly, you match up business/functional/technical requirements, the support team’s capabilities, and the best products & technology to form a solution in the most cost effective manner. Of course you want to try and put in the best possible option, and it is easy to focus strictly on the technology. But, cost must always be a consideration. At the end of the day, you either make money for the business or you cost money to the business, and you need to help that equation no matter which side of it you are on. In network infrastructure, you are often a cost center. Granted, we need to run the business in a robust manner, but you will have a hard time justifying a Ferrari to your boss when a Honda Civic will do the job just fine.

Of course all new designs are a balancing act between these goals. But, if you can achieve these three for an infrastructure design, you are in good shape! Plus, you’ll get more sleep, and your project might have a little more funding for a steak dinner.

Droid X: The best smart phone (for now...)

Ever since the announcement of the Android mobile operating system, I have pined for an Android phone. I am a strong supporter of open source, and I have not been crazy about the various smart phone platforms available.

Roughly 1.5 years ago, I upgraded my phone from a personal plain phone to a smart phone funded by my employer. To sign up for this program, I had to pick a Windows-based smart phone per employer standards. This was indeed an upgrade, as I now had email, contacts, and calendaring synced from my work accounts. The phone did what I needed for work, but the platform itself was limiting. It was clear Windows Mobile 6.1 was designed by translating desktop elements to a mobile device, and that translation did not work. The concepts of "apps" for a mobile device was new (and arguably is still new for MS), and the phone was underpowered.

Meanwhile, Android started gaining an immense amount of momentum. Google released their own phone. Verizon, Google, and Motorola partnered to release the Droid 1 (along with much marketing). Many handset makers started releasing Android phones on multiple carriers. The app market grew from a few hundred to nearly 80,000 apps, and many mainstream app developers now release Android versions as well as iPhone.

Finally in the summer of 2010, two critical efforts pushed me over the edge. First, HTC and Google both added two key components to their Android versions: the ability for an Exchange admin to require a PIN such that the phone is locked, and the ability to send a "remote wipe" signal to the phone to delete all data. These features are both required by my employer, and this enabled me to seriously start considering an Android phone. Google also added these features in their 2.2 version.

Secondly, my company-provided phone started having technical issues. It could not reliably send MMS (picture texts) messages. The stars aligned, and I was able to upgrade to a Droid X.

The Droid X is an amazing piece of technology. Co-workers have joked that its more powerful than the first laptops we were issued roughly 13 years ago, and that it is like a super computer in my pocket. Admittedly I had very high expectations after watching the industry and platform for so long, and this phone meets most of them!

The combination of Motorola hardware, Android 2.2, and Verizon's excellent wireless service is very impressive. Motorola's wireless division has "bet the farm" on Android, and it appears to be paying off and resulting in some great products. While I won't be giving up a laptop anytime soon, this phone is truly an excellent mobile device. The typical smart phone functions (email, calendar, phone) work great, are fast, and are easy to read on the large screen. The library of applications is truly impressive for both work and personal use. I thought the size might be a bit unwieldy. It is only a tad larger than my previous phone, is nice and thin, and fits great in my pocket and on my ear for calls.

I'm still in the "honeymoon" gadget period, and I think this honeymoon will last a while.

Now admittedly Android is not the best phone for everyone. While I have some philosophical issues with Apple, they do make some amazing hardware and products. If you just want your phone to work, are not into "gadgets," don't mind Apple's walled garden, and want a smart phone, the iPhone may be a good choice. But if you want to see the future of smart phones, like to experiment/tweak/play with gadgets a bit, and don't want some CEO telling you what you can and cannot load on your mini-computer, get a Droid X.

Of course, Android phones are moving along at an amazing pace. I will be interested to see what comes out this holiday season. There is some talk about multi-core CPU's soon, HDMI out will likely be a typical feature, and Verizon's LTE network is coming soon. It is a very interesting time for this market.