Section 3 describes the Data Architecture and Application Software entities of the DODIIS Profile. Section 3.1 discusses the DODIIS Data Architecture. Depending on the type of functionality they provide, DODIIS applications are characterized as Support Applications or Intelligence Applications, which are described in sections 3.2 and 3.3, respectively.
In the existing environment, numerous databases, containing structured and unstructured (free-format) data, graphics (such as maps, overlays, and annotations), and imagery, are available to intelligence analysts. However, there are several major shortcomings to the methods currently used for storing, representing, and providing data to support intelligence analysis. Among the most prominent are:
The DODIIS objective data handling capability is to present analysts with a comprehensive, consistent view of multimedia data. Various items of data, regardless of medium, will be linked together according to predefined relationships included in a data dictionary/directory, and presented to the analyst as the result of a single query. The analyst will not have to know where the data is located, or the mechanisms needed for retrieving it. The data may be located in a single database or contained in physically separated computers that use different database management systems managed and controlled by different organizations. Further, only one version of any given data element will exist, although there may be multiple copies distributed in such a way as to minimize retrieval time.
In addition, the information system will support "views" into a global data model. Analysts (and applications) will be able to specify relevant subsets of the total data available. Imagery analysts could, for example, use a predefined "Imagery Data" view. The analyst could also specify data categories for a tailored view composed of a combination of imagery, relevant free-text messages, broadcast television news reports, and maps. A key point of the objective data architecture is that analyst views into DODIIS data could be changed readily to meet the needs of the analyst without significant modification of the underlying data structures.
Within this context, a DODIIS Data Architecture is required that allocates elements of data to specific data managers, and appropriate sets of data to user communities or sites. On the one hand, responsibility for maintenance of each element of data will be assigned to a particular organization. On the other hand, each element of data will be made available to all analysts that have the appropriate need-to-know and access privileges. Thus, while the responsibility for update is centrally located, other copies of the data in existence at DODIIS sites will be updated automatically from the "master" copy. Since only one master copy will exist, data redundancy will be controlled, minimizing the possibility that multiple versions of a single data element will exist. (Note that any copy of an element of data could be slightly dated at any given time, depending on the mechanisms used for replicating data.) Management of this distributed data concept will be accommodated through centrally controlled, normalized data dictionaries. Optimally, the distribution of data will be achieved with a distributed database architecture and software design that minimize the need to duplicate data elements from master data dictionaries.
The DODIIS Data Architecture does not imply that site-unique data will not exist. The requirement for unique data at each site will be accommodated to support site-unique missions and functions through local extensions to the master data dictionary.
Further, site databases may contain "shadow" or command data elements to facilitate crisis management or to account for unique collection and/or processing factors that create the potential for data synchronization/integrity problems. Analysts, however, will not have to know where the applications and the data reside; they will have transparent access to various functions and the data required to complete their tasks, through the distributed processing capabilities inherent in information systems designed in accordance with the DODIIS Profile.
The Stored Data portion of the DODIIS Profile represents the actual data used by the intelligence analyst. While the DODIIS Profile identifies the need for the DODIIS community to develop a global view of the data produced and used by the community, where it is stored, how and by whom it is maintained, and how the distributed data is kept current and concurrent is not addressed fully in this document. The DODIIS Data Architecture will be produced in a separate effort under the auspices of the DODIIS Management Board (DMB).
DOD Directive 8320.1, Data Administration, establishes a policy for DOD data administration; authorizes the establishment of and assigns responsibilities for a DOD data administrator to plan, manage, and regulate data within the DOD; and authorizes the establishment of a DOD information resources dictionary system. In accordance with this Directive, the Principal Deputy for Intelligence within the Office of Assistant Secretary of Defense has been designated functional data administrator (FDAd) for the defense intelligence portion of the DOD data standardization process. The responsibilities of the defense intelligence FDAd include overseeing the development of standard intelligence data models, proposing data element standards for the intelligence community, identifying industry standards that are applicable to the intelligence community, and coordinating intelligence activities with other DOD FDAds.
The primary DOD data administration goals are to develop a centrally controlled DOD-wide data repository to: 1) manage and store standard data definitions, formats, and structures (such as architectures and subject area models); 2) standardize data elements, data models, and data architectures to facilitate data sharing, data quality, reuse, and integration; and 3) develop common procedures and tools to support functional managers, data administrators, and the technical support community. To achieve these DOD objectives, management commitment is needed at levels. DODIIS commitment to data standardization is evidenced by its practices over the last several years. For example, DODIIS has undertaken initiatives to standardize the data elements the intelligence community uses to define order of battle intelligence and perform collection tasking. Work initiated by the DIA in the area of data element standardization (i.e., the Intelligence Data Element Authorized Standards [IDEAS] program) has been adopted as a DODIIS community initiative and submitted for consideration as an initial baseline for the DOD information resources dictionary system (i.e., for inclusion in the Defense Data Repository System [DDRS]). Ensuring that DODIIS migration systems (see section 3.3 ) support IDEAS will be the first move toward DODIIS compliance with DOD standardization guidelines. As new data standards are proposed and specified through the DOD data standardization initiative, they will be evaluated for applicability to the intelligence community and, when it is appropriate, integrated into DODIIS migration systems.
Support applications are tools that provide DODIIS personnel with automation support to assist in completing their mission. Support applications provide automated support in the DODIIS environment by two means:
These two categories do not provide any specific differences in support applications other than to distinguish the level of distribution and availability in the DODIIS environment. Common support applications are part of the DODIIS infrastructure and should be readily available on any DODIIS workstation.
DODIIS is in the process of selecting a suite of standard support applications that are to be used for all future migration systems and DODIIS sites through a standard support contract, namely the System Acquisition and Services Support (SASS) Contract. All products selected for DODIIS through SASS will be mandatory for all DODIIS sites and Migration Systems. All SASS specified DODIIS products are denoted with bold text in table 3-1. Table 3-1 identifies a representative set of the most common types of support applications found in the DODIIS environment. Most of the support applications listed in Table 3-1 are listed by functionality and are not intended to identify specific COTS or GOTS products.
All DODIIS support applications must comply with applicable standards identified in section 2 . In addition, there are installation and development goals for the selection of DODIIS support applications:
As previously mentioned these are goals and should be used as evaluation criteria when selecting candidates or developing support applications in the DODIIS environment. The overall DODIIS goal is to provide the maximum use of standard COTS products in the DODIIS automation environment. COTS products are the primary focus of the DODIIS support applications environment, but there are times when cost, implementation or availability will mandate the use of GOTS or unique developed products.
DODIIS projects must provide DODIIS management with technical and cost trade-off analyses of all functions that could be implemented using COTS products when GOTS or developed products are being considered. Final decisions on buy or build will depend on the salient cost and technical issues.
In order to contain the functional costs of performing the DOD mission, the Deputy Secretary of Defense directed, in a 13 October 1993 Memorandum, that DOD Components focus their efforts on functional process improvement, selection of standard migration systems, and data standardization objectives that are an integral part of the Corporate Information Management initiative. In particular DOD Components were directed to accelerate the selection and deployment of migration systems and the selection and implementation of data standards. Near-term guidelines established in the 13 October memorandum were augmented by the ASD(C3I) in 12 November 1993 and 20 December 1993 memorandums. The Intelligence Systems Board (ISB) evaluated and consolidated the inputs for intelligence community migration systems.
As a first step in complying with the 13 October DOD Memorandum, the DODIIS Management Board (DMB) directed the formation of a joint Engineering Review Board (ERB) and System Integration Management Office (SIMO) team. The team's objective was to select DODIIS Migration Systems and submit the nominations to the Intelligence Systems Board (ISB) for approval. By-products of the DODIIS Migration Systems selection process included the identification of community legacy systems and the recognition that there is a need to support site-unique systems within the overall migration effort.
DODIIS functional managers performed a high level business process analysis to identify current and future mission needs and validate the analytic processes used to achieve intelligence objectives. Using the validated business process analysis as a guide, the functional managers categorized the functional activities performed as part of the intelligence process. The categories agreed upon are based on those defined in Joint Publications 2-0 and 2-01:
Migration Systems were selected for each of the categories and/or sub-categories. Details of the selection process can be found in a Defense Intelligence Agency (DIA)/S-FM Memorandum, subject DODIIS Migration Systems. The transition to the target architecture (i.e., an architecture in which all DODIIS sites are solely supported by DODIIS Migration Systems and site-unique systems) is expected to be completed by 31 March 1997.