CHAOS ^arc : kernel support for multiweight objects, invocations, and atomicity in real-time multiprocessor applications

Abstract

CHAOS arc is an object-based multiprocessor operating system kernel that provides primitives with which programmers may easily construct objects of differing types and object invocations of differing semantics, targeting multiprocessor systems, and real-time applications. The CHAOS arc can guarantee desired performance and functionality levels of selected computations in real-time applications. Such guarantees can be made despite possible uncertainty in execution environments by allowing programs to adapt in performance and functionality to varying operating conditions. This paper reviews the primitives offered by CHAOS arc and demonstrates how the required elements of the CHAOS arc real-time kernel are constructed with those primitives.

I-RMI: Performance Isolation in Information Flow Applications

“…There is a plethora of work addressing runtime performance management in distributed server systems, ranging from system-level solutions like process/load migration or request throttling [26,27,32], to application-level tradeoffs in the quality of server responses produced for clients vs. server response time [10,18], to the creation of new middleware or system abstractions that support the runtime adaptation of applications and systems in response to changes in user requirements or platform resources [7,13,15,20,24,25].…”

“…We monitor a server's steady state throughput (see Fig. 2) using request counting methods similar to [13]. When a sudden drop in throughput (L 1 ) and at the same time, a sharp increase in resource utilization is detected, we identify a potential 'poison' state.…”

Invocation of real-time objects in a CAN bus-system

“…CHAOS [ [11]] demonstrates that the object model can be tailored to meet the efficiency requirements of realtime applications. The motivation also comes from a robotics application where electromechanical components are encapsulated and controlled by objects.…”

“…Although the trade-off generally is between adaptability and predictability, these systems are scalable in the sense that they allow a coexistence of activities with different real-time requirements. These systems rely on run-time guarantees [[30]], [ [11]] rather than on a preplanned feasibility guarantee only. A form of calendar-based scheduling [ [22]], [[30]] is an approach to run-time guarantees where hard real-time activities have the possibility to reserve resources in advance.…”

Roo: a framework for real-time threads

“…As an example, imagine the new scheduler uses priorities in order to implement a more advanced scheduling scheme where threads are moved between different priority queues depending on their run-time behaviour 11 . Since the basic prio class o ers the method move taking a thread and its new priority a s parameters and moving the thread from the priority queue it is currently on to the priority queue denoted by the new priority, all the programmer has to do is design the new class to decide, based on the given policy, when to call the move method of its baseclass prio.…”