THE DISTRIBUTED COMPUTATION IN ACTIVE NETWORK

 

Phan Xuân Huy

University of Natural Sciences - VNU.HCM

Abstract:

 

          Traditional data networks passively transport bits from one end system to another. Ideally, the user data is transferred opaquely. The network is insensitive to the bits it carries and they are transferred between end systems without modification. The role of computations within such networks is extremely limited, e.g., header processing in packet-switched networks. Active network allows their users to inject customized programs into the nodes of networks. An extreme case replaces traditional passive packets with "capsules" - program fragments that are executed at each network router/switch they traverse. These network are "active" in two ways: (1) Switches can perform complex computations on the user data flowing through them. (2) Individuals can inject programs into the networks, thereby tailoring the node processing to be user and application-specific. Theirs advantages have been well identified: they provide task distribution over the network, permit a massive increase in the sophistication of the computation that is performed within the network. This paper makes an attempts to give an overview of active network architecture, some approaches to its design, preliminary ideas in exploiting an active network framework for distributed computation.