8:30 AM discussion with <vendor deleted> about pricing
10:00 AM Review of two projects to be expedited
10:30 AM (double booked with above) Traffic Filtering discussion
11:00 AM Follow-up discussion with <vendor deleted> (competing with vener above)
1:00 PM D2405 Project Status Meeting
2:00 PM EVDO timer meeting
Okay... that's the scheduled meetings--about 6 hours worth. Now I have 2 really hard hitting projects right now they are numbered D2405 and D0840. D2405 is huge, like mega-huge. It involved setting up a new IP data service between Sprint and all other carriers. The number of possible combinations is large and each possible combination has to be documented in detail with network diagrams and flow diagrams. Each object on those diagrams has to have details about where it is, how many interfaces it has (each interface has to have where it goes and it's own particulars), how the device will be managed, how the device will be monitored, what will happen when each of it possible failures occurs, and how much it costs. Each device also needs to be compared to other devices so that the "correct" one can be selected. Every possible system failure must be documented and the microsecond steps of what happens documented. As well as "logical" diagrams (of what device connects to what) you have to have "physical diagrams" show exactly how the devices will fit in to a rack. You have to capture power requirements and cooling requirements (Remember for *each* device).
On top of all the techincal documentation you then have to tie every single technical decision to a "Business Requirment". Like "Knives were required to deliver on the Sliced Bread requirement." There are literally hundreds of business requirements...
You can't just buy equipment. Vendors have to slog it out in a battle of contracts. Who ever can deliver the most product for the least cost wins. Sounds good, but it takes months of testing and negotiation. All the while you have to... guess... document all the tests and results and empirically compare them so that you can prove to your VP (no technical ability at all) that your technical decision is sound.
After all the vendors are selected, you have to build a replica of everything in a lab. You then test every possible use case and failure that you documented. You have to, of course, document all the elements of the lab used for testing and all the results.
Once testing is complete, you can deploy your first production test site. Again with the documentation, testing and results.
Once all these simple tasks are complete you're done and you move on to the next project.
I have two of these going on at the same time...
I'm currently trying to document the first production test of D0840 and I'm trying to get the vendor out of negotiation for D2405. I'm waaaaay behind on the documentation.
Between the 6 hours of meetings I have to get all the documentation done. I only have 2.5 days this week since I'll be flying out to Flordia for a conference on Wednesday. The conference is a meeting of all the carriers to discuss the new IP data product and how we're all going to interconnect. 90% of the attentees will be speaking english as their second (or possible third) language. All of them will have a PhD in something (I don't). All of them are going to decide to spend millions of dollars on using an intermediary rather than save millions by choosing to connect to some directly.
So there's the work load. I also have to keep up with my own Sprint paperwork (performance reviews, etc). My house still isn't unpacked, my gutters need to be cleaned (or my basement is going to start leaking), my truck needs to be washed, I really want to finish the decor in my kids' rooms and the rest of the house... I'm just wondering if I ever really get to catch up.
No comments:
Post a Comment