Many application domains (e.g., avionics, telecommunications. and mult
imedia) require real-time guarantees from the underlying networks, ope
rating systems, and middleware components to achieve their quality of
service (QoS) requirements. In addition to providing end-to-end QoS gu
arantees, applications in these domains must be flexible and reusable.
Requirements for flexibility and reusability motivate the use of obje
ct-oriented middleware like the Common Object Request Broker Architect
ure (CORBA). However, the performance of current CORBA implementations
is not yet suited for hard real time systems (e.g., avionics) and con
strained latency systems (e.g., teleconferencing). This article descri
bes the architectural features and optimizations required to develop r
eal-time ORE end systems that can deliver end-to-end QoS guarantees to
applications. While some operating systems, networks, and protocols n
ow support real-time scheduling, they do not provide integrated soluti
ons. The main thrust of this article is that advances in real-time dis
tributed object computing can be achieved only by systematically pinpo
inting performance bottlenecks; optimizing the performance of networks
, ORE end systems, common services, and applications; and simultaneous
ly integrating techniques and tools that simplify application developm
ent.