Computing Infrastructure Readiness (CIR)

The DoD net-centric vision requires information and services to be visible, accessible, understandable and trusted across the Department. Information is an enterprise asset, decoupled from associated applications, and ready and accessible to meet previously unanticipated needs arising from new circumstances or mission needs. Today’s environment is typified by dedicated hardware for specific applications. Information is tied to the application, the location and the operating system. Current Defense Enterprise Computing Centers (DECCs) and equivalent non-government implementations focus on determining capacity and utilization requirements for each individual system or application. This approach can lead to poor utilization of computing resources and require additional hardware and software to be purchased to accommodate dynamic usage and future growth. Contingency planning in this paradigm is accomplished by reserving capacity dedicated to a specific use.

As the Department moves farther down the net-centric operations path, the underlying computing infrastructure for core applications and services must transition towards the delivery as a net-centric capability and not discrete chunks of technology. Net-centric CI will leverage the GIG’s distributed computing resources to provide infrastructure that appears to the end-users or applications as one virtual capability. Shared computing and data storage resources will be virtually allocated and the mechanism for doing so will be transparent to users. By “virtualizing” how users view the computing infrastructure, DoD can begin reducing technical and administrative barriers to sharing resources, and provide more agile and scalable support for information sharing across the GIG.

As computing infrastructure evolves to better support net-centric operations, it must take into account the needs of edge users – those at the forward or leading edge of the mission operations environment. The concept of building and maintaining an agile computing environment must support end-users operating in environments challenged by intermittency and low bandwidth.

The CIR priority focuses on the Department's challenges in transforming its legacy of system-specific computing infrastructures to shared GIG computing infrastructure nodes that can deliver guaranteed levels of capability to consumers and providers of the Department's data and services. CIR seeks to transform DoD GIG CI from a hardware- and program-centric infrastructure, to one that is dynamic, shared, adaptable and sufficient to support global net-centric operations.

Delivering Net-Centric Computing Infrastructure

Computing infrastructure in the net-centric environment will be customer-driven, shared, dynamically allocated and automatically monitored and configured. Net-centric computing infrastructure will enable:

  • Location-independent storage - Services and applications will share storage anywhere across the GIG, allowing consolidation and efficient use of data storage resources. Likewise, users will be able to access information transparently from anywhere across the GIG.
  • Dynamic, automated storage provisioning - Experience-derived knowledge and use patterns will be used to heuristically allocate data storage. Thus, CI will be able to "learn" from past usage experience to better serve users.
  • Virtualized application environments - Applications will be hosted in shared versus dedicated environments, enabling dynamic changes to processor and storage capabilities depending upon usage patterns. Hosting environments provide seamless access to all applications and services regardless of their physical location.
  • Automated status reporting - All GIG CI resources will continually report their status, thus enabling NetOps to have a continuous view of the status of CI resources across the GIG for situational awareness and command and control monitoring.

Transitioning to this net-centric computing infrastructure will bring many benefits to the Department's operations:

  • Reduced complexity - Many of the Department's existing capabilities have grown through independent acquisitions of components, without an overall vision or architecture in mind. The emerging best practices for large-scale data center operations (including management by SLAs) will drive simpler, more consistent infrastructures.
  • Better responsiveness - The ability to monitor GIG infrastructure across all applications, services, and user groups, along with the ability to respond dynamically to data storage and processing load will make it easier, faster and less expensive to allocate additional resources to meet new, unanticipated demands.
  • Shared CI Resources - With the ability to share resources dynamically among applications, services, and user groups, peak transient CI demand for an application or service can be met by prioritization of the CI "pool" and by providing available infrastructure resources dynamically in response to priority uses.
  • Increased consolidation opportunities - With the ability to share processing and storage, the need to build excess capacity in every individual application’s hardware in order to meet increased or unexpected user demand will be eliminated. This will have a dramatic positive effect on the overall cost of operations.
  • Support to the edge- The focus on highly available and accessible information resources that scale dynamically to meet user demand and geared to support continuous operations will greatly enhance capabilities available to users at the forward edge of the mission operations environment.

Standardizing GIG Computing Infrastructure Nodes

In the net-centric operating environment, applications and services will no longer be hosted and maintained on dedicated hardware. They will be resourced virtually on GIG Computing Nodes (GCNs) spread across the GIG’s pooled resources. GCNs are IT facilities that provide hosting for applications and services, data storage and content staging in controlled environments that ensure capabilities are delivered within specified service levels. GCNs provide managed physical security, backup and Information Assurance capabilities for all IT services they host. As depicted in the table below, GCNs may be established at several different scales, ranging from fixed enterprise scale computing centers, to regional or area processing centers, down to local- or unit-level computing centers. Standardizing definitions and rules around GCNs is essential to delivering net-centric computing infrastructure capabilities successfully. GIG CI resources must be brought to the edge via a robust, responsive, and adaptable fixed and deployed GCN taxonomy. All GCNs must be IA / NetOps certified and accredited for adherence to computing service provider (CSP) adequacy criteria.

 

The principles and rules detailed on the next page should guide the Department in building agile computing infrastructure environments that will support net-centric implementation of IT applications and services, meeting the needs of all to the edge.