You are currently viewing the printable version of this article, to return to the normal page, please click here.

Army rolls out brass to defend anti-IED software

Makes its case against Palantir

Question of the Day

Is it still considered bad form to talk politics during a social gathering?

View results

The Army presented two two-star generals and three intelligence specialists Thursday to defend its $2.5 billion battlefield intelligence processor, which has failed operational tests and has been criticized by soldiers as being too slow to analyze the enemy and help find buried bombs in Afghanistan.

In a news conference at the Pentagon, the Army specialists lauded the Defense Common Ground System, an array of computers, servers and programs that is the Army's principal processor of huge amounts of battlefield data.

The Washington Times first reported in July about an internal battle within the Army. Commanders and intelligence officers in Afghanistan complained in messages to Army headquarters about the Defense Common Ground System.

Some asked for permission to buy Palantir, an off-the-shelf software platform that specializes in linking disparate bits of information to form a clear picture of the battlefield.

In some cases, Army officials involved in shielding the Defense Common Ground System from possible budget cuts viewed Palantir as a competitor and worked to shut off the requests.

Thursday's news conference offered a different view.

The generals said the Defense Common Ground System, which has been in development for a decade, grew out of a giant gap in intelligence collection: There was no single database to bring together information in Afghanistan and look for links among suspected enemy fighters.

"We had a difficult time [in] what we were collecting and even what it was collecting, when it was collecting and where it was collecting," said Maj. Gen. Stephen Fogarty, who heads the Army Intelligence and Security Command. He has held senior posts in special operations, Central Command and Afghanistan.

"Data from all that collection resided in different databases that were often incompatible," Gen. Fogarty said.

A different tale

Maj. Gen. Harold J. Greene, the Army's deputy for acquisition and systems management, said the Defense Common Ground System "replaced nine different intelligence systems."

"This really is a change in methodology," Gen. Greene said. "It reduces the amount of hardware we have to buy and the footprint on the ground."

Intelligence processing has become especially important in fighting terrorists such as the Taliban, who wear no uniforms and hide among the population, making them difficult to identify.

"It's really the Army intelligence-analyst weapon system," Gen. Fogarty said. "DCGS has been used effectively in Iraq, Afghanistan and around the world by both conventional and our special operations forces. We believe it enhances soldiers' situational awareness and improves commanders' ability to protect the force."

The Times has obtained a series of messages from combatants that tell a far different story.

Officers lauded Palantir for its ability to zero in on data that helped soldiers find roadside bombs, the No. 1 killer of troops in Afghanistan. They complained that the Army-issued Defense Common Ground System was too slow.

A November 2011 memo from the 82nd Airborne Division provides an example:

"Solving very hard analytical problems takes several days when using existing tools against these data sources," the message states. "In our experience in using the Palantir platform against the same problems, we were able to reduce this time to a few hours. This shortfall translates into operational opportunities missed and unnecessary risk to force."

'People have preferences'

Rep. Duncan Hunter, a California Republican and a former Marine officer who served in Iraq and Afghanistan, has accused the Army of fighting requests for Palantir to protect its own system.

Mr. Hunter said of Thursday's news conference: "The Army's in damage-control mode. DCGS has underperformed in many areas and created capability gaps that soldiers are looking to fill with alternative solutions.

"The Army can talk about intent all it wants, but there's a difference between what the program's intended to do and how it's actually performing. Otherwise, there wouldn't have been at least 13 separate requests from ground combat units for something different," said Mr. Hunter, a member of the House Armed Services Committee.

The generals said nine of those requests were granted, and gave various reasons why the other four were rejected.

"Every intelligence officer wants the best capability available," Gen. Fogarty said. "People have preferences. We have some very aggressive analysts. Frankly, they drive us to continue to improve the system."

He asserted that once a brigade opts to use Palantir "that data is not completely available, is not interoperable" with other intelligence systems.

"The ease of use [with Palantir], that has been very important to them," he said.

An aide for Mr. Hunter said the nine-of-13 statistic does not tell the full story: Commanders went around the Army bureaucracy to get approval from other agencies or waited months for an OK when they needed Palantir right away.

Failing grades

The generals said the Army is conducting an industry competition among Palantir and other software providers to determine whether their link-analysis software can be incorporated into the Defense Common Ground System.

Col. Dave Pendall, a former division intelligence officer in Afghanistan, told reporters that the Defense Common Ground System performed well in his eastern region. He used it to produce daily intelligence reports on data from human sources, intercepts of electronic transmissions, satellite images and other sources.

"In my experience, DCGS has met our requirements," Col. Pendall said.

Both the Army's and Pentagon's chief testers have issued failing grades for the Defense Common Ground System's latest version.

The Defense Department's Operational Test and Evaluation office said the system was "not operationally effective, not operationally suitable and not operationally survivable against cyberthreats," according to a copy of the evaluation obtained by The Times.

Gen. Greene said the tests show "we had some challenges with the work flows" and "some reliability challenges."

The Pentagon's top acquisition official last week agreed to let the Army field the latest version of the system, but without software to store top-secret material. That so-called "enclave" was causing systemwide problems. Without it, the newer system should be able to perform as well as the ones now in-country, the Army says.

The Senate Armed Services Committee report on fiscal 2013 defense spending scolded the Army for refusing to approve commercially available systems such as Palantir.

"The Marine Corps and even some Army units in Afghanistan proceeded to deploy commercial products," the report said, referring to the fact that officers bypassed the Army hierarchy to buy them. "Overall, the feedback from these units and an independent assessment by the deputy secretary of defense-chartered National Assessment Group has been very positive on these commercial products. Unfortunately, the Army's cloud analyst support appears to continue to lag behind promised performance."

© Copyright 2014 The Washington Times, LLC. Click here for reprint permission.

Comments
blog comments powered by Disqus
TWT Video Picks