Utility (On Demand) Computing

The concept of utility, or on demand, computing is new, but it represents a logical next step in the evolution of IT services in the 21st century. Utility computing is an approach to the use of computing resources—hardware, software and networks—the object of which is to maximize the efficiency of the use of these very expensive assets. The following features typify a utility computing implementation:

• A standardized architecture using a specific processor, specified operating system(s), and an integrated and centrally-managed Storage Area Network;
  
• Architecturally-defined and separate environments for development, testing and production;
  
• Standardized support processes using industry best practices (e.g., IT Infrastructure Library, or ITIL);
  
• Operations follow established Service Level Agreements (SLAs) for security, availability, disaster recovery, break/fix and network processes; and
  
• The network model provides all necessary network connectivity, computing platforms, storage facilities, security processes and procedures, disaster recovery processes and procedures, and Federal standards compliance.

Under a utility computing model, an the IT assets of an enterprise are managed centrally and as a whole, eliminating the inefficiency of redundancy across business units but compensating those units for the loss of their autonomy with access to resources (speed, storage, applications, etc.) that individual business units could never afford on their own.