[dba-SQLServer] My Relative Absence Here

Arthur Fuller artful at rogers.com
Thu Feb 9 19:49:09 CST 2006


I just peeked and there are numerous unread messages from this group in my
inbox. I do try to pay attention to this list, but I just got a new gig and
subsequently got tossed into a situation where there are serious deadlines
impending, serious enough that I have been given permission to work Saturday
and Sunday too. The database is what is known as large rather than rich: a
rich database has many hundreds of tables containing relatively few rows; a
large database has many millions of rows in relatively few tables. This one
is medium-rich (200+ tables) and fairly large (some tables contain 50
million rows). To further complicate matters, "the database" is actually
comprised of about a dozen databases, each residing on its own server, so
all joins typically contain a reference to the server as well as the
table(s) of interest. 
I am charged with double-checking the Erwin PDM model against the Data
Modeller's LDM and comments on the last printout. 779 comments and one week
to do it. 
When I got the gig, I received a file containing 279 issues, and I expressed
confidence that I could do it in a week, though overtime might be required.
Several days later, I was informed that I had been sent an old version of
the Issues file; the current version contains 779 issues.
I replied, "I can tell you with 100% confidence that I will not meet the
deadline. 779 issues / 40 hours = 19.475 issues resolved per hour; each
issue resolved in 3.15 minutes, on average. Many issues (i.e. are the
indexes on this table enough, or redundant, or incorrectly designed?) will
require a meeting, which will consume a minimum of 10 minutes, if we are
lucky and there are no politics. Some will require an hour or two to settle.
So my confidence in my inability to meet the one-week deadline is 100%. You
have two chances: a fat one and a slim one."
They said, "Monday?"
I said, "Maybe. No guarantees, but I will work overtime and do my best to
make it happen. Think Tuesday or maybe Wednesday and maybe even next Friday.
There are consequences to failing to meet the deadline: not for me
personally, but for the whole team. So I am trying my best. I have to read
the critique, then modify the Erwin physical model to suit, then inform the
DD (Database Design) guys that column X has become NOT NULL, etc., or that
index Y has been removed due to redundancy, or that it's time for another
meeting with the DM (Data Modellers) and this time let's not come to blows.
All this, and I have to commute and dress business-casual, too. If it wasn't
for the filthy lucre, I'd rather be doing something else. That isn't true! I
love this kind of thing! Thrown off the pier and expected to swim
immediately. It's a thrill.
I am on a team of very smart people, each of whom has an expertise in a SQL
area that pales my own knowledge of that area. The data architect guy is
brilliant and has written a book about SQL that proves it (Google Dejan
Sunderic for a list of his books). The DTS guy is brilliant. The sprocs guy
is brilliant. The manager guy is brilliant. The Data Modeller guy is
brilliant, and to make it more interesting, he and the Data Architect have
frequent arguments.
I'm neck deep in intelligence. As poker players say, "Look around the table
and if you don't see the fool, you're it." I look around the table and
conclude that I am the dumb guy. This week alone I learned about 10 really
cool things, some about SQL Server itself, some about Erwin, some about pure
database design.
I just hope that I am able to demonstrate enough intelligence to keep up
with these boys. They KNOW their stuff. And I hope to come close enough to
the arbitrarily imposed deadline to retain the gig and not be fired for
incompetence. The opposite of the poker player's perspective is, "What a
pleasure it is to be surrounded by brilliant players."
---------------------
So that, in short, is why I haven't been paying my usual attention to this
list. I have a TON of work to do in a very short time, and it requires a lot
more concentration than I have been accustomed to shelling out of late:)
Most important is to complete the task at hand. By that time, I may have 100
emails to visit on this thread alone.
All I wanted to say, I guess, is that I still consider myself a member of
this list, and I will catch up next week, maybe even as soon as Monday.
Arthur




More information about the dba-SQLServer mailing list