Everyone agrees that modernization of
federal IT infrastructure and systems is necessary, and the swift
passage of the Modernizing Government Technology Act will help. But one
area of disagreement is the viability of “legacy” technology and what
exactly should be done about it.
The government has been reporting that 75 to 80 percent of the
federal IT budget is spent on running legacy (or existing) systems. That
may sound alarming to those who aren’t familiar with the inner-workings
of a large IT organization. However, the percentage is in-line with the
industry average. Gartner says the average distribution of IT spending
between run, grow and transform activities — across all industries — is
70 percent, 19 percent and 11 percent respectively. Those numbers have
been consistent over the past decade.
In addition, the mere fact that a technology system is legacy, or
old, does not mean that it isn’t effective. When it comes to working
code, the longer it’s been running, the better it becomes as bugs and
inefficiencies are eliminated over time. CIOs from the Department of
Health and Human Services, the IRS and the Defense Department are just a
few that have said that their legacy systems, written in Assembler and
COBOL, are well written and can be kept current through ongoing
stewardship.
Unfortunately, some technology leaders look past these facts and
advocate an overall “rip, rewrite and replace” strategy. In my nearly 30
years with major IT companies, I’ve seen how blanket replacement
strategies and new, unproven technologies have fared. More often than
you might expect, the result is a waste of precious time, money and
energy on an outcome that doesn’t achieve the stated goals. Consider
Pennsylvania’s lawsuit against IBM, where a major modernization project
to deal with its legacy technology resulted in complete failure.
So, with $500 million in taxpayer dollars ready to be spent, it’s
imperative that we take a look at legacy modernization approaches that
have worked and those that have failed.
Cloud-first mentality
The cloud is not the answer to every technological need.
A blanket re-platforming of core legacy applications is an enormously
expensive and risky undertaking. The key is understanding when the
cloud is appropriate and when other platforms offer the optimal
approach. Large private organizations with the same “can’t fail”
reliability needs as the government have long depended on the mainframe
for their systems of record. Is their mainframe code “legacy?” Well, it
may have started decades ago, but the hardware and software have been
continually updated. Ninety-two of the world’s top 100 banks still use
mainframes and wouldn’t think of migrating off the platform.
Cloud-first has become a misguided proxy for what really matters,
which is continuously improving the citizen experience in engaging
federal services. This is the perfect time for federal CIOs to think
deeply and critically about a cloud-first strategy, and ask themselves
if they are really doing what’s in the best interest of U.S. citizens or
wasting time and taxpayer dollars chasing the “next big thing.” This
fundamental question should drive every federal CIO’s vision and
mission.
The two platform approach
What is working well is a hybrid approach we call two platform IT,
which smartly uses the best of both worlds in appropriate situations.
Part one takes advantage of the mainframe’s unmatched security,
reliability and performance-at-scale for transaction processing and
other systems of record — for example, the U.S. federal tax system. Part
two uses the cloud for other services or applications that, while
essential, don’t require the near-perfect reliability the mainframe
offers – for example, HR applications.
The mainframe modernization issue
It is true that the experienced mainframe workforce is primarily made
up of Baby Boomers who are retiring. What’s not true is that this
workforce needs to be replaced with employees of equal experience. A new
generation of post-modern mainframe solutions bring agility and ease of
use to the mainframe. This includes everything from updating the
mainframe interface, thereby making it familiar to new IT workers, to
speeding up the development processes supporting the mainframe, which
has not traditionally been known for rapid development cycles. The
result is a multi-platform system that can keep pace with the demands of
our mobile, digital economy.
It’s also important to note that a well-managed and maintained
mainframe is also the most secure computing infrastructure available,
requiring much less protection from outside attack. If you’re thinking
“What about the massive OPM breach? Wasn’t that a mainframe problem?”
Actually, no. The distributed systems surrounding the mainframe allowed
the breach, and if simple, available security measures had been taken,
it could have been avoided. The post-modern mainframe requires 69
percent less effort to secure than other systems,
according to IBM research.
So as the MGT’s new funding and flexible budgeting opportunities are
clarified in the coming months, federal CIOs will have to interpret the
law to best leverage their legacy systems and determine what changes
will truly serve the citizens paying for them.
Is it old and in need of replacing? Or is it a proven technology that
simply requires modernization? MGT dollars will be best spent with
careful consideration to these questions.
Chris O’Malley is CEO of Compuware,
where he is responsible for setting the company’s vision, mission and
strategy. With 30 plus years of IT experience, Chris has led the
company’s transformation into becoming the “mainframe software partner
for the next 50 years.”
