Evaluating policies and mechanisms to support distributed real‐time applications with CORBA

Abstract

To be an effective platform for performance-sensitive realtime systems, commodity-off-the-shelf (COTS)

Multiparadigm scheduling for distributed real-time embedded computing

“…To better meet these requirements, researchers at Washington University St. Louis and the University of California, Irvine have developed a second-generation ORB called TAO [8], which is an open-source implementation of Real-time CORBA that supports efficient, predictable, and flexible DRE computing. Prior work on TAO has explored many dimensions of high-performance and real-time ORB design and performance, including scalable event processing [9], request demultiplexing [10], I/O subsystem [11] and protocol [12] integration, connection architectures [13], asynchronous [14] and synchronous [15] concurrent request processing, adaptive load balancing [16], meta-programming mechanisms [17], and IDL stub/skeleton optimizations [18].…”

Patterns and performance of distributed real-time and embedded publisher/subscriber architectures

“…TAO is an opensource 1 implementation of Real-time CORBA that supports efficient, predictable, and flexible DRE applications. Our prior work on TAO has explored many dimensions of highperformance and real-time ORB design and performance, including optimal request demultiplexing [12], I/O subsystem [13] and protocol [14] integration, connection architectures [15], asynchronous [16] and synchronous [17] concurrent request processing, adaptive load balancing [18] and meta-programming mechanisms [19], and IDL stub/skeleton optimizations [20].…”

Techniques for enhancing real-time CORBA quality of service

“…Likewise, our work on efficient, scalable, and, predictable dispatching components [23] shows how it is possible to bound priority inversion and hence provide timeliness guarantees to DRE applications. Moreover, our work on Real-time CORBA explicit binding mechanisms illustrates how to design predictable connection management into an ORB [9].…”

“…Our prior work on Real-time CORBA has explored many dimensions of ORB design and performance, including scalable event processing [4], request demultiplexing [5], I/O subsystem [6] and protocol [7] integration, connection management [8] and explicit binding [9] architectures, asynchronous [10] and synchronous [11] concurrent request processing, and IDL stub/skeleton optimizations [12]. In this paper, we consider how to achieve end-to-end predictability using Real-time CORBA.…”