I haven’t been really talking about the fact that I was pursuing my VMware VCDX certification over the last 6 months or so. It’s a long process that takes a lot of work and I didn’t want to mention it in case it didn’t end well. I’ve been working on this post over the last week and just saw that Chris wrote up his experience here. A lot of what he says mirrors exactly what I was going to recommend. Read his post..listen to what he says, especially about the application and the documents.
The start of this process was to take the Enterprise Admin and Design exams. Those are now retired so there isn’t much to say about them. One thing I will note is that these are taken at Prometric Professional centers, not your standard Vue testing center. The difference here is that there aren’t nearly as many of those and scheduling can become a real problem. I barely got my Design exam scheduled before the deadline to defend at VMworld 2010. If I had failed it I wouldn’t be able to retake it for at least a month, and that’s with me looking at all centers within 500 miles. So plan ahead! Get on the schedule for these! Don’t wait until the last minute. The deadlines happen well before you expect them so email the VCDX group and get the list of dates for your milestones based on when you plan to defend.
Once you finish those two exams you’re maybe 1/3 of the way there. The real fun starts when you receive your application for your design. It’s not a big document, but the beast lies within. Consider this document the “Table of Contents” for the rest of your design work. It should point to those other deliverables and provide a good detailed overview of the design. I get asked a lot how long it took me to create my design and I get to give my favorite answer…it depends. Are you doing a design from scratch? Are you modifying an existing design? Is it a real design or a theoretical design? All of these can make the time vary greatly. One person I talked to said it took them 9 days working on nothing but their design. That’s pretty much how long it took me. I worked on mine one full weekend, all through the week, and then through the second weekend and finally turned it in right at the deadline on Monday. I would have liked to spend a little more time on my design but unfortunately I had a family emergency mid-week that I had to attend to..but I estimate that 9 or 10 full 10 hour days is a very good estimate.
While the application does spell out many things that are required it also leaves a lot to interpretation. I consider this a good thing. Different people document designs different ways. Many VMware partners use the Plan & Design kit as their base framework but that does NOT make a complete design deliverable. Don’t think you can just use those templates and you’re set. Other people may want to just create their templates from scratch or use internal documentation. It’s up to you. The key, though, is to make sure everything in the application is covered. John Arrasjid, VCDX 001, will sometimes Tweet some good tips and information on the VCDX process, especially the application. Duncan Epping chronicles those here. Read them. Follow them. Why wouldn’t you listen to John’s advice? By the way, he’s also a great guitar player as many of us saw at the VMware party on Wednesday. Nice, John.
In the end you want to create a well thought out design. Don’t just focus on the technical details but also the “softer” parts of the design such as how you deploy, manage, and maintain the environment. Are you considering how a customer manages the environment you’ve designed after you left, or just worrying about how it looks on the last day of implementation?
Assuming you finish everything on time you then submit your bundle of documents to VMware and pay $300 for them to review. If everything goes well you’ll be invited to defend in front of the VCDX panel. Now the real fun begins…. My defense was scheduled for the Tuesday before VMworld 2010. I thought that was kind of odd as I was expecting the defenses to occur during VMworld but after talking to some people they said the logistics just didn’t work out. Defending a week before VMworld meant I had to fly to San Francisco a week early which greatly adds to the expense of the trip. I figured I’d make the best of the situation and had my wife fly out on Wednesday for a vacation. We wanted to extend my trip to Las Vegas for Cisco Live but couldn’t do to the aforementioned family emergency so we were long overdue. That was a nice 4 days and helped get my mind over what happened next….
The defense. First things first. There isn’t a lot you need to do to prepare for the defense. You need to create a presentation to use as an introduction and overview of your design. Remember, the panel will have all of your documentation at their disposal during the defense. The presentation given at Partner Exchange, and available here, talks about the presentation as well as all parts of the defense and gives some good examples. The idea is that it’s a short, maybe 15 minute, presentation to get you started and calm your nerves. I asked for advice on how deep to go in the presentation. Some people do a long presentation that goes in to detail, others do a short quick run through the design. Again, also read the tips from John that I linked earlier as he discussed the presentation. The key here is you don’t want the presentation to bog down the defense. You get 75 minutes to cover it all. If you don’t get to something you may not receive any credit for those pieces which is a big problem. Keep that clock in mind.
To say I was stressed before my defense is a huge understatement. Ask anyone that really works with me and they’ll tell you that it takes a lot to rattle me. I can speak in front of a few hundred people and not break a sweat. Things like that just don’t phase me…but this defense…yikes. My primary goal that day was not to throw up before (or I guess, during) my defense. The morning of the event I got up and grabbed a light breakfast….I didn’t feel like eating but I knew not eating was far worse. I figured the panel would deduct points if I started hallucinating. After breakfast I headed down to Palo Alto and hung out at a Starbucks going over my presentation and design. When it was time I rolled over to the VMware campus. Chris walked around the campus but I didn’t. I went ahead and checked in 20 minutes before my start time and waited a while…and then a while more…then I started to get concerned as I didn’t want to get there late and lose any of that precious time…finally someone came to get me and took me to the defense room. It’s an interesting feeling walking in that door. Good? Bad? I’m not sure. Good in that the work up to that point has paid off. Bad in that the work up to that point hangs on what happens in the next couple hours.
The panelists greeted me and were very friendly. We talked a bit as I set up my notebook to get ready for my overview presentation. And then we’re off…. My presentation was about 20 slides and instead of 15 minutes it went well over that, but it was due to the panel asking many questions throughout. This part of the defense was enjoyable. Once we got started the stress went away and it was fun. It was intense. It was stressful. There were times where my mind was racing…but it was fun. I know not everyone enjoys things like that but it’s akin to an adrenaline rush. The best advice I can give you here is the same advice everyone gives. Know your design. Know it well. Study it. I don’t care that you wrote it yourself. I don’t care if you implemented it at a customer. There is a big gap between when you turned in the application and when you are in front of that panel. If you know something you can answer a question quickly and easily. If you have to refer to another document you’re wasting time. Time is key. You get 75 minutes. You don’t get 75 minutes and 1 second. Beyond just knowing the design know WHY you made the choices you made as well as what else those choices impacted. Duncan has a great post here on Impact of Decisions. Read it.
After your defense you get a 10 minute break… Use this time to relax and clear your head. Get a drink of water. Go to the restroom. Look out a window. Whatever… But don’t waste the time.
When you get back in “the room” things change and you’re no longer focusing on your design. For the next 30 minutes you’ll be doing an ad hoc design with the panel as customers. I can’t go in to any detail or give any information on the type of scenario or how it was presented. If you read that VCDX presentation linked earlier they give a sample of what you can expect to see. This shouldn’t be too bad, right? I do this every day. I’m a pre-sales architect!
I cratered. It was ugly. I never got a complete train of thought going. It was a plane wreck in slow motion in my head. It was like I could hear myself yelling “Pull up! Pull up!”. “Abort! Abort!”. Before I could regain control my time was up….and the ad hoc was over with only a small fraction of the things I needed to cover done. I would give a lot to be able to just repeat those 30 minutes, but that’s the point of the exercise, right?
I fully expect this to cause me to fail the defense. No one outside the panel knows how things are scored but I know it can’t be good. Everyone says they are sure I’m overreacting, but I was there. 😉 It’s disappointing since this is the piece that I expected to be able to manage but maybe that’s their plan. I have a feeling I’m the perfect use case for the candidate to flail. This part was far tougher than I had anticipated. How often do you do a customer design or even gather requirements in 30 minutes with zero knowledge of the customer? It flustered me. I figure this is a taste of what it’s like to be a pro athlete and choke during the big game at a big moment. All the work leads up to that point and then….it’s over.
The final piece of the panel is the 15-minute troubleshooting section. The idea here is to show how you can tackle a problem. The best advice I can give here is to think out loud. Talk. Even if you’re talking in circles. It gives the panel insight in to how you are tackling the problem. I had to keep reminding myself to do that. This section is similar to the ad hoc design. Until they put the scenario in front of you there is no way to know what to expect. In my opinion….you don’t know what you don’t know, though right?….this went pretty well. I solved the problem with about 2 minutes left though I’m not sure I took the best technical path. Does that matter? Who knows, except for the panel. The problem with these types of tests…at least for me….is that I try to think of how to do these things in front of a panel. You don’t want to just jump to a conclusion and end up wrong. So my thought process was to just break things down in to pieces and work the issue. Hopefully that came across. Don’t worry if you do this part and don’t solve the problem. Ask current VCDXs and I bet you find many didn’t get there in the allotted time. Solving it isn’t the problem…it’s how you tackle problem.
So now I wait. I think the fourth part of the defense is second guessing every single thing after. Like I said..I’m pretty sure I know the outcome of this. It’s disappointing but I’m already working on how to handle things differently the next time. My primary concern right now is the actual VCDX track moving to VCDX4. I have an email in to the certification team asking about that. I took the VCDX3 exams, which are now retired. Will I be able to resubmit a design based on the VCDX3 and hopefully defend at like Partner Exchange or will I have to first take the new exams, effectively starting over? I hope not…but we’ll see. When I find out the results, I’ll let you know….