Let's
tie together the last several columns on "2012 Might Really be
The End of the World." In this series, When discussed several
megatrends in the general IT industry that will have a tremendous
impact on the database administration (DBA) profession. The
megatrends include both software-related (virtualization and cheap
cloud database services) and hardware-related (SSDs and massively
multi-core CPUs). These technologies have the potential to
obviate many of the core competencies of the DBA, with the first two
eliminating or lessening the need for server and hardware
configuration and provisioning, and the last two diminishing the need
for IO tuning and query tuning, respectively. But those are
trends that will take years to reach fruition. What about the
near future?
In
actuality, prospects for current and future DBAs. Here are a
few reasons.
First,
most companies, even many Fortune 500-sized organizations, aren't
getting basic database administration right in the first place.
They can't institutionalize the benefits of basic information
management techniques until they've had a few years of solid
experience with their data. These organizations, both great and
small, need some time to understand their database before they can
exploit their data assets or effectively extract business
intelligence from them. I'm doubtful that immature
organizations can leapfrog into the newest technologies without
wasting a lot of money, time, and morale.
Second,
there aren't enough DBAs, and there won't be for the foreseeable
future. The Bureau of Labor Statistics shows there are
currently 380,000 people working under the title of "databaseadministrator." (The most recent data I could find was
from 2009.) Of course, many more do the work of a DBA without that
title. However, database administration and its variants are
considered high-growth. There's only about a 4 % unemployment
rate for DBAs, and demand far outstrips supply.
Third,
our higher education system just doesn't get it. It's not
turning out the kind of graduates needed to step immediately into
database management professions. Even the most savvy and
progressive of universities don't think deeply about data.
They're still focused primarily on algorithms (for the Computer
Science department) or requirements and business analysis (for
Information Systems majors). And those who do think about data
usually offer, at best, two courses on database management.
So,
a new graduate must spend at least a few years in the trenches before
he or she can become an effective part of a data management team.
This effectively winnows out many potential candidates. It's
also given rise to the term "accidental DBA," which we see
daily in the SQL Server world on blogs and conference presentations.
The accidental DBA is someone who was given DBA responsibilities with
no formal training on the subject. He or she might be the best
SQL developer on the dev team. Or they might be the most
analytical of the business analysts. Sometimes, they just
happen to be the person standing closest to the server rack when the
boss says, "We need a volunteer to be a DBA - ah, you!"
With
so many accidental DBAs put in charge of valuable data assets,
organizations essentially start the cycle all over again, which goes
back to my first point - companies aren't getting even the basics of
database administration right.
So,
yes - if you're a DBA today, you have some things to learn and some
things to worry about in the future. But you have plenty of
time to learn them. Once you learn new technologies like
virtualization and IO tuning with SSDs, you'll be able to leverage
those skills into promotions and pay raises. Because as long as
we have "accidental DBAs" coming up from the trenches,
large organizations will have to pay a premium for the elite data
management professionals who know both the best practices for
operational database administration - and how to take advantage
of the newest technologies.
No comments:
Post a Comment
Thank You For Comment...
Your comment is under moderation wait for approval