Defense

Army personnel database picks up where Pentagon left off

file folders on a background with binary code

After years of foundering efforts by the Pentagon to create a department-wide personnel and pay system, the individual services have assumed the task. In the Army's case, the consolidation of 1.1 million soldiers' records amounts to one of the largest data transfers in military history.

The Integrated Personnel and Pay System-Army (IPPS-A) offers information on soldiers' pay, assignments, awards, benefits and promotions, among other data. The database consolidates more than 40 disparate human resources systems by either interfacing with or subsuming them.

Despite the potential for pitfalls when handling that much data, it has been a smooth ride so far, according to Col. Robert McVay, the man in charge of the project.

"From my perspective, the biggest positive of [the database] is that for the first time, we have standardized the source and the data for every soldier's record," said McVay, who has been IPPS-A project manager since October 2011. He is in the process of handing IPPS-A off to Col. James "Darby" McNulty in what an Army spokesperson said was a standard rotation of program managers.

The Army's new pay and personnel system grew out of the Pentagon's decision to cancel a department-wide database known as the Defense Integrated Military Human Resource System (DIMHRS). DOD reportedly spent $1 billion on the program that Adm. Mike Mullen, then chairman of the Joint Chiefs, called a "disaster" in 2010 because of its poor performance and excess cost.

The Army is keen to draw a line in the sand after DIMHRS: an answer to a "frequently asked question" on the IPPS-A website debunks the notion that the new database is simply DIMHRS reincarnated. "Instead of DIMHRS's large 'Big Bang' approach, IPPS-A is deployed incrementally with testing, training and implementation focused by H.R. and pay function," the online memo says.

Military services are not implementing their databases in isolation. McVay said he meets at least quarterly with IPPS program managers from the Air Force and Navy to share project experiences and identify challenges ahead. "We have not been proprietary with our database," he said.

Shared learning between the services in implementing IPPS databases goes back to at least when McVay took on the project two and a half years ago. The Coast Guard had recently finished up its database and offered advice to McVay in leading the Army's development.

The great migration of data across Army systems to IPPS-A is being facilitated by Oracle's PeopleSoft software, which was also used in the ill-fated DIMHRS project. McVay said the latest version of PeopleSoft has been honed for a better fit to IPPS-A's purposes.

Compared to DIMHRS, the new Army database is being rapidly implemented. Deployment began March 25 and by April 30 IPPS-A was available to the National Guard. In the coming weeks it will be rolled out to active-duty and reserve soldiers.

Despite the progress, McVay senses there are some in DOD who would like to see the deployment speed up. But that, the colonel said, would be a mistake. "When you're dealing with … thousands of transactions that have to go on in the life of a soldier, going faster isn't necessarily a good thing," he said.

The consensus from big data gurus when consolidating multiple data sources is to first "harvest all of the data from all of your systems, normalize it and put it into one database," McVay said. With the data centralized, the Army can make more complex queries of soldiers' data by whatever variable it chooses. That level of specificity was not possible in the old H.R. systems.

The 2014 Federal 100

Get to know the 100 women and men honored this year for going above and beyond in federal IT.

Reader comments

Tue, May 6, 2014

It's about time! Wish it would have been completed before I retired.

Mon, May 5, 2014

Col McVey is dead on when he advises: "harvest all of the data from all of your systems, normalize it and put it into one database" but he left out one other important step in the data transformation: "sanitize it!"

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above