How should DCGS-A approach its big data challenges?
The U.S. Army's Distributed Common Ground System-Army (DCGS-A) enables Army units to collect and consolidate data from each unit's internal sources, plus over 700 external sources. (1) DCGS-A then merges and fuses the data, thus establishing relationships between the data to help leaders make decisions. During this process each Army unit employing DCGS-A may process terabytes (TB), if not petabytes (PB) of data during a military operation. How then should DCGS-A approach the challenges of dealing with a potentially unmanageable amount of data? Big Data is one of DCGS-A's core functions insofar as it directly impacts mission success or failure, life or death, and victory or defeat.
This article examines DCGS-A's Big Data challenges. It starts with an overview of the DCGS-A system and its intended use, and then continues by defining Big Data in the context of the DCGS-A program. Next, the article analyzes the DCGS-A Big Data strengths, weakness, opportunities, and threats. The article then concludes with an exploration of possible big data solutions for DCGS-A, to include, but not limited to: hardware and software tools, data storage and analysis service, and non-materiel solutions as prescribed by the Army (doctrine, organization, training, leadership, personnel and facilities).
What is DCG5-A?
The DCGS-A Public Affairs Office website characterizes DCGS-A as a system that consolidates battlefield data obtained by Soldiers and sensors from national, theater, and tactical level assets.
It then analyzes these vast amounts of data and provides decision makers an enhanced picture of the enemy and battlefield conditions. This "picture" (consisting of graphic, image, and text products) is commonly referred to as a common operational picture (COP). It provides leaders with situational awareness and enables Army units to "see first, understand first, act first, and finish decisively".
The DCGS-A Program Manager (PM), in cooperation with the Training and Doctrine Command's Capability Manager-Foundation, is responsible for designing, developing, fielding, and sustaining DCGS-A. The PM's strategy separates DCGS-A into two increments. Increment 1 (the current system) is comprised of:
* Fixed nodes in sanctuary locations not on the battlefield, providing data and services to the mobile nodes.
* Mobile nodes embedded with units on the battlefield, consisting of servers and workstations utilized in tents and/or vehicles (see Figure 1), that receive data and access services to process data.
* The Army network that interconnects the fixed and mobile nodes.
Increment 2 (the future system) will build upon Increment 1 by adding more services, as well as the Cloud-based network architecture (depicted in Figure 2), that provides users access to data and services from anywhere in the world.
What is Big Data in the Context of DCGS-A?
Big Data is a massive amount of organized and/or unorganized data. (4) Big Data is characterized by its volume, velocity, variety, value, and vera city. (5,6) Army units (using the DCGS-A) receive data from their internal and location specific sources in addition to over 700 external sources. (7) This equates to data volumes ranging from terabytes to petabytes during a military operation. Furthermore, military operations often have a high operational tempo meaning the velocity of data received for analysis will likewise increase exponentially. This data will include a wide variety of structured data (such as pre-formatted reports that are readily incorporated into a database) and unstructured data (e.g., images and graphics).
Along with the volume, velocity, and variety of data, Army units must possess the ability to discern the veracity (e.g., accuracy, usefulness, and reliability) of the information. In other words, which items are "garbage in" as they will invariably produce "garbage out" results. Finally, the value of the collective data is influenced by each of the previous attributes, and derived from each unit's utilization of DCGS-A to analyze the data. For the Army, value is ultimately determined by the extent to which data provides situational awareness and facilitates decision making.
How Does Big Data Impact the DCGS-A Program?
The volume, velocity, variety, plus veracity variance of data processed by Army units highlight some of DCGS-A's internal strengths and weaknesses. They also present DCGS-A with various external opportunities and threats. Figure 3 summarizes the DCGS-A strengths, weaknesses, opportunities, and threats (SWOT).
The first Big Data strength for DCGS-A Increment 2 is its status as a "new start" program. Increment 2 possesses the flexibility to incorporate new technologies and operating procedures into the system, to include advances in Cloud Computing and Big Data management and analysis. This dovetails well with the first opportunity, the availability of new/upgraded Big Data and Cloud Computing technologies that were not available to Increment 1. The combination of this strength and opportunity will enable DCGS-A Increment 2 to address some of its Big Data challenges.
The second strength, availability of terabytes to petabytes of data, will provide units exponentially more data from which to draw conclusions. However, this can also lead to the Big Data weakness of data overload, resulting in units receiving more data than they can process in a given time period. Army and Department of Defense investments and technological advances in sensors that increase the volume, variety, and velocity of data received by Army units further compounds the weakness of data overload. This paradigm shift is a proverbial double-edged sword for DCGS-A. Increases in data provide units with more historical and real-time data that improves the fidelity of their trend analysis; conversely, more data elevates the "messiness" of each unit's overall data set.
"Messy" data includes errors and "inexactitudes" of data. (8) For example, two separate sensors may report on (supposedly) the same entity within a short time of one another. Any deltas in the sensors' reporting are errors (e.g., sensors are looking at different entities) or inexactitudes (e.g., sensors are reporting on the same entity, but from different perspectives, thus producing somewhat differing reports). However, authors Mayer Schonberger and Cukier (see Endnote 8) believe that despite the risk of elevated messiness, more data is always better.
For DCGS-A this is true when conducting trend analysis in an effort to predict what an opponent will do in the future. More data increases the confidence in the correlation between indicators (shaping actions) that precede major events (decisive actions). As Mayer-Schonberger and Cukier point out, for predictions it is less important to understand why an opponent does something, and more important to understand what indicators the opponent will display prior to conducting decisive action.
SWOT: DCGS-A Inc 2 (Big Data) Internal Factors External Influences Strengths Opportunities + Initial Increment (Inc) 2 in 2015, + New technologies available to as a "new start" program; Inc 2 that were not available and/or affordable for Inc 1: > insert new technologies and > Big Data procedures > Cloud Computing + Inc 2 will access TBs to PBs of + Potentially more funding under data during a unit's deployment: a new Army Civilian Executive leadership in Jan 2017 > more Real-time data for immediate action > more Historical data for tend analysis * see below: increased data is an Opportunity and Threat Weakness Threats - Data Overload: * Investments and technological > Inc 1 had challenges with advances in Sensors are managing the plethora of increasing the volume, variety, data produced/received and velocity of data > Inc 2 must manage more data than - Increases in data creation also Inc 1 increase security and privacy concerns - Inc 1 heavily reliant on - Near total reliance on Army point-to-point network and DoD network for Cloud connectivity between nodes versus Architecture untested Inc 2 Cloud architecture - Backwards compatibility with Inc 1 Figure 3. DCGS-A SWOT Matrix.
On the other hand, messy data can introduce an unacceptable level of risk. Military operations frequently include life-or-death situations; thus, making critical decisions based upon data considered "messy" is generally not acceptable. These situations require higher fidelity (e.g., "eyes on") realtime data, confirmed by multiple sources, prior to taking action. For example, before sending troops to take an objective by force, units (utilizing their access to real-time data) can obtain confirmation (and reconfirmation) of their opponent's status via multiple independent sources. This is one of the great benefits of Big Data for Army units, and DCGS-A Increment 2 will further enhance this capability by sorting through the "messy" data.
How Will DCGS-A Handle its Big Data Challenges and Threats?
The DCGS-A PM Office, plus Army units utilizing the system, can exercise various steps to maximize DCGS-A Increment 2's strengths and opportunities, while simultaneously minimizing its weaknesses and threats. First, as previously mentioned, the PM Office can insert new and/or enhanced Big Data technology into the system. For instance:
* Simultaneous transaction processing (user interaction) and analytic processing (trend discovery and pattern identification) using the same system. For DCGS-A this could reduce the number of workstations required in each unit by combining multiple functions into a single integrated platform.
* Search and interactive analysis of structured data through a new visualization interface. For DCGS-A the visualization interface could enhance the human-to-machine interface and make it easier for users and decision makers to understand the data and results of analysis.
* A query approach that enables analysis of unstructured data on systems, such as Hadoop. ("Hadoop is a free, Java-based programming framework that supports the processing of large data sets in a distributed computing environment.") (9) This tool could enhance DCGS-A's ability to process the plethora of unstructured data it receives, such as images and graphics.
* Visualization and exploration of Big Data that samples and profiles data automatically to create catalogs (organized listing of metadata). This solution could bolster DCGS-A's organization of data, especially unstructured data, and increase user's ability to find, understand, and utilize data.
The second step for the DCGS-A program concerns cloud computing. With the proper cybersecurity protections, this could enhance DCGS-A's data storage capacity as well as Army units' access to analytical services. Cloud computing is examined in more depth in a separate article.
Army units and users must also take actions that, under Army parlance, are non-materiel solutions including changes in doctrine, organization, training, leadership, personnel, and facilities. A paramount priority is to establish full-time, dedicated, and properly trained knowledge managers who are responsible for ensuring that the data DCGS-A uses, as well as the results and application of DCGS-A's analysis (i.e., information and knowledge), are properly managed (e.g., metadata-tagged, discoverable, and accessible).
Furthermore, but of no less importance, units must create, adopt, and enforce cybersecurity policies and procedures to protect data from hostile forces. Finally, Army leaders and decision makers must understand what Big Data and its analysis can and cannot deliver, especially in real-time life and death situations.
Big Data is critical to the U. S. Army. Data are the building blocks for the DCGS-A, and DCGS-A Increment 2 will provide Army units the tools to manage the complexities of Big Data. These complexities include messy vs. unmessy data, structured vs. unstructured data, and the five "Vs" of Big Data: volume, velocity, variety, and veracity; all of which collectively affect a unit's ability to draw value from the internal, local, and externally produced terabytes to petabytes of data that a unit must handle during a military operation.
DCGS-A Increment 2 will require both materiel and non-materiel capabilities in order to effectively and efficiently manage Big Data. For materiel solutions, the system will require data handling and analysis capabilities by Oracle: Oracle 12c, Oracle Business Intelligence Enterprise Edition, Big Data SQL, Big Data Discovery, and Business Intelligence Cloud Service. For the non-materiel solutions, the Army should institute knowledge managers for each unit employing DCGS-A, establish and enforce cybersecurity policies and procedures to protect data, and promote a firm understanding of the benefits and limitations of Big Data.
Using DCGS-A Increment 2, Army units will have the ability to differentiate the vital data from the interesting but less (or not) relevant data, connect-the-dots between the volumes of Big Data at their disposal, and form a picture of the operational battlefield environment and activities that convey a shared understanding of the situation. This COP is the key product of the DCGS-A data analysis, the situational understanding it provides is the goal, and enhancing leaders' decision making capacity is the ultimate objective.
by Lieutenant Colonel (Ret.) Jake Crawford
LTC(Ret.) Crawford served 23 years on active duty in the Army Acquisition, Military Intelligence, and Adjutant General Corps, and deployed to Iraq and Afghanistan. For his final assignment he was the Army Test and Evaluation Command System Team Chair for DCGS-A, responsible for the developmental and operational test and evaluation of the DCGS-A Increment 1 system. He is a graduate of the U.S. Military Academy at West Point, earned an MBA and is currently enrolled in an MS program for Information Technology and Systems Engineering.
Terabyte - A terabyte (TB) is a large allocation of data storage capacity applied most often to bard disk drives. Hard disk drives are essential to computer systems, as they store the operating system, programs, files and data necessary to make the computer work. Depending on what type of storage is being measured, it can be equal to either 1,000 gigabytes (GB) or 1,024 GB. Disk storage is usually measured as the first, while processor storage os the second.
Petabyte--A petabyte (PB) is an even larger allocation of data storage capacity. As with terabytes (TB), depending on what type of storage is being measured, a PB can be equal to either 1,000 TB or 1,024 TB.
(1.) Colonel Robert M. Collins, "DCGS-A Inc 2: The Evolving Environment and Transition to Open Competition," 2014. At http://www.afcea.org/events/armyintel/14/documents/COL_Collins_2014.pdf, 3.
(2.) "FY 2014 Annual Report: Distributed Common Ground System-Army (DCGS-A)", Office of the Director, Operational Test & Evaluation, 20 January 2015. At http://www.dote.osd.mil/index.html.
(3.) "DCGS-A", DCGS-A Public Affairs Office, 7 May 2014. At http://dcgsa.apg.army.mil.
(4.) Efraim Turban, Linda Volonino, and Gregory R. Wood, Information Technology for Management: Advancing Sustainable, Profitable Business Growth (9th ed.) (Hoboken: John Wiley & Sons, Inc., 2013), 7.
(5.) "About Big Data", University of Maryland University College, 2015. At http://www.umuc.edu/analytics/about/big-data.cfm.
(6.) Paulo B. Goes, "Big Data and IS Research," MIS Quarterly 38(3) (2014): iii-viii.
(7.) Collins, 3.
(8.) Viktor Mayer-Schonberger and Kenneth Cukier, Big Data: A Revolution that Will Transform How We Live, Work, and Think (New York: Houghton Mifflin Harcourt Publishing Company, 2013).
(9.) "Definition--Hadoop", Tech Target.com. (n.d.). At http://searchcloudcom puting.techtarget.com/definition/Hadoop.
|Printer friendly Cite/link Email Feedback|
|Publication:||Military Intelligence Professional Bulletin|
|Date:||Oct 1, 2016|
|Previous Article:||Capitalizing on situational awareness geospatially enabled tools: reflections following a RAF rotation.|
|Next Article:||DCGS-A and the data management challenges.|