I/O-Intensive Embedded Systems: The Infopipe Approach
Abstract
Infopipes are a distributed computational and communications abstraction for information flow applications and I/O intensive distributed real-time embedded (DRE) systems. Infopipes are specified by the syntax, semantics, and quality of service requirements for information flows. Software tools generate executable code from the specification. Building applications that process information flows on existing middleware platforms is difficult, because of the variety of QoS requirements, the need for application-specific protocols, and the poor match of the commonly used abstraction of remote invocations to streaming. The ability to query individual Infopipe elements as well as composite Infopipes for properties of supported flows enables QoS-aware configuration. Similar to local protocol frameworks, Infopipes provide a flexible infrastructure for configuring communication services from modules, but unlike protocols the abstraction uniformly includes the entire pipeline from source to sink, possibly across process and node boundaries.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 2004
- Accession Number
- ADA435430
Entities
People
- Carlton Pu
- Greg Eisenhauer
- Jonathan Walpole
- Karsten Schwan
- Ling Liu
- Mustaque Ahamad
- Wenchang Yan
Organizations
- Georgia Tech Research Corporation