AStreams An approach to adaptive distributed systems

Systems Group
College of Computing
Georgia Institute of Technology


[ Project overview | People | Publications | Code | Related work | Contact information ]

Project overview

An increasing number of distributed applications aim to provide services to users by interacting with a correspondingly growing set of data-intensive network services. Such applications, as well as the services they utilize, are generally expected to handle dynamically varying demand on resources and to run in large, heterogeneous, and dynamic environments, where the availability of resources cannot be guaranteed a priori - all of this while providing acceptable levels of performance.

To meet this challenge we propose a middleware approach that we call Active Streams. Actives Streams, and its associated framework, are intended to explore the idea of application-level distributed adaptation. Through Active Streams, users can dynamically build/extend applications and services by attaching location-independent functional units to the data streams flowing between their applications and network services. The C-based framework employs dynamic code generation in order to insure that these functional units can be executed efficiently, without the overheads of an interpreted language; and provides for fine-grain adaptation of their behavior through the association of read-only blocks of data whose content can be updated remotely in a push-type operation. Finally, these functional units can be dynamically re-deployed over the datapath in order to best leverage the available resources in the face of the dynamically changing environment and the stress of ever-changing application behavior.

People

Code distribution

Active Streams is under development. Some parts of the infrastructure it relies on can be obtained from:

Publications

Related work

Infosphere InfoSphere

Research projects and papers on topics related to Active Streams.

Contact information

If you have additional questions about this research, please feel free to contact us.


Fabian E. Bustamante
Systems Research Group
College of Computing
Georgia Institute of Technology
Atlanta, GA 30332-0340
(404) 385-2022

fabianb@cc.gatech.edu