| R. Rajkumar and L. Sha and J. P. Lehoczky, "Real-Time Synchronization Protocols for Multiprocessors ", In IEEE Real-Time Systems Symposium, pages 259-269, 1988. |
....scheduling algorithms is that only independent tasks that do not synchronize or share resources have been considered. In contrast, tasks in real systems usually are not independent. Synchronization entails additional overhead, which must be taken into account when determining system feasibility [2, 6, 18, 19, 20, 22]. Unfortunately, prior work on real time synchronization has been directed at uniprocessor systems, or systems implemented using non fair scheduling algorithms (or both) and thus cannot be directly applied in fair scheduled multiprocessor systems. Indeed, synchronization issues in fair scheduled ....
R. Rajkumar, L. Sha, and J. Lehoczky. Real-time synchronization protocols for multiprocessors. In Proceedings of the Ninth IEEE Real-Time Systems Symposium, pages 259--269. 1988.
....heuristics is a non trivial problem. Further, such constraints also make the uniprocessor schedulability test less tight, and hence, partitioning less e#ective. The multiprocessor priority ceiling protocol (MPCP) has been proposed as a means to synchronize access to resources under partitioning [33]. Unfortunately, the MPCP was proposed only for RMscheduled systems and needs to be adapted for use in EDF scheduled systems. To the best of our knowledge, no multiprocessor synchronization protocols have been developed for partitioned systems with EDF (though it is probably not di#cult to extend ....
R. Rajkumar, L. Sha, and J. Lehoczky. Real-time synchronization protocols for multiprocessors. In Proc. of the 9th IEEE Real-Time Systems Symposium, pages 259--269. IEEE, 1988.
....the amount of time a high priority process must wait for a lock to be freed depends not only on the duration of some critical sections, but possibly on the duration of the complete execution of some processes. However, with the use of priority inheritance this time is bounded. It is described in [Rajkumar88] how, if we have m lower priority processes holding onto k distinct locks, the maximum number of critical sections that a real time process will have to wait is at most MIN(m, k) 3.5 Internal Event Processing Callout processing is the traditional UNIX way of handling internal timeout events. In ....
. Rajkumar, L. Sha, and J. P. Lehoczky, "Real-Time Synchronization Protocols for Multiprocessors", Proceedings of the Real-Time Systems Symposium, Dec. 6-8, 1988, Huntsville, Alabama.
....standard, portable way to ensure consistency between the internal synchronization mechanisms used by an ORB and an application. For realtime applications, however, it s necessary to ensure consistency between synchronization mechanisms to enforce priority inheritance and priority ceiling protocols [20]. Therefore, the Real time CORBA specification defines a set of locality constrained mutex operations that ensure consistency between synchronizers used by the ORB and its applications. 5 Concluding Remarks DOC middleware is a promising paradigm for decreasing the cost and improving the quality ....
R. Rajkumar, L. Sha, and J. P. Lehoczky, "Real-Time Synchronization Protocols for Multiprocessors," in Proceedings of the Real-Time Systems Symposium, (Huntsville, Alabama), December 1988.
....Lack of QoS enforcement: Conventional ORBs do not provide end to end QoS enforcement, i.e. from applicationto application across a network. For instance, most ORBs transmit, schedule, and dispatch client requests in FIFO order. However, FIFO strategies can yield unbounded priority inversions [52, 53], which occur when a lower priority request blocks the execution of a higher priority request for an indefinite period. Likewise, conventional ORBs do not allow applications to specify the priority of threads that processes requests. Lack of performance optimizations: Conventional ORB endsystems ....
R. Rajkumar, L. Sha, and J. P. Lehoczky, "Real-Time Synchronization Protocols for Multiprocessors," in Proceedings of the Real-Time Systems Symposium, (Huntsville, Alabama), pp. 259--269, December 1988.
.... applications to (1) determine the priority at which CORBA invocations will be processed, 2) allow servers to pre define pools of threads, 3) bound the priority of ORB threads, and (4) ensure that intra process thread synchronizers have consistent semantics in order to minimize priority inversion [28]. It is important to recognize that RT CORBA s priority mechanisms cannot work miracles. In particular, ORB middleware cannot magically imbue a non real time OS or communication infrastructure with completely deterministic behavior [29] When used in the appropriate environment, however, certain ....
....server declared model is useful for certain real time applications, it is not suited for all application use cases. For instance, one way for a server to avoid priority inversions is to process incoming requests at a priority equivalent to the client thread that invoked the operation originally [28]. The RT CORBA client propagated model allows clients to declare invocation priorities that must be honored by servers. In this model, each invocation carries the CORBA priority of the operation in the service context list that is tunneled with its GIOP request. Each ORB endsystem along the ....
[Article contains additional citation context not shown here]
R. Rajkumar, L. Sha, and J. P. Lehoczky, "Real-Time Synchronization Protocols for Multiprocessors," in Proceedings of the Real-Time Systems Symposium, (Huntsville, Alabama), December 1988.
....strategies for improving the 5 performance and predictability of database transactions: 1. use memory based databases [9, 11, 37, 41] 2. schedule transactions according to task priorities or deadlines [1, 6, 33, 43, 42] 3. reduce delays and uncertainties associated with concurrency control [6, 9, 13, 15, 20, 30, 31, 35, 36, 39, 48]. To this list, we add: 4. avoid the overhead associated with a client server architecture, and 5. run transactions in parallel on shared memory multiprocessors. Some real time database systems have been developed for main memory which can dramatically enhance performance and predictability by ....
R. Rajkumar, L. Sha, and J. P. Lehoczky, "Real-time synchronization protocols for multiprocessors, " in Proc. Real-Time Systems Symposium, pp. 259--269, December 1988.
....2.2.1. In other words, there exists systems which are schedulable with PIP but not with PCP. The opposite is also true: As chained blocking cannot occur with PCP, some systems that are schedulable with PCP cannot be scheduled under PIP. MULTIPROCESSOR PROTOCOLS. Rajkumar and associates [RSL88] proposed an extension of the PCP for multiprocessor systems, called the multiprocessor priorityceiling protocol (MPCP) In that protocol, all resources are bound to fixed processors. The protocol distinguishes between local resources and global resources. Local resources can be accessed from ....
R. Rajkumar, L. Sha, and J. P. Lehoczky. Real-time synchronization protocols for multiprocessors. In Proceedings of the 19th IEEE Real-Time Systems Symposium, pages 259--269, 1988.
....code, however RT CORBA still lacks higher level synchronization mechanisms such as condition variables, semaphores, and barriers. Moreover, as no standard API for the priority inversion protocol is prescribed by RT CORBA, the user must rely on ORB specific APIs to fully prevent priority inversion [15]. An additional advanced feature exploited in the proposed framework is drawn from the CORBA 2.4 Messaging specification [13] Standard service requests in CORBA systems rely on the Synchronous Method Invocation (SMI) model, that blocks the client until the server notifies the end of the ....
R. Rajkumar, L. Sha, and J. P. Lehoczky. Realtime Synchronization Protocols for Multiprocessors. In IEEE Real-Time Systems Symposium, Huntsville, AL, December 1988.
....approach is to consider the task including an I 0 blocking as the two subtasks, one before the blocking and one after it. For testing the schedulability of lower priority tasks than the task ro which includes an I 0 blocking, the inva sive effects of deferred execution must be considered [5]. Specifically, the subtask of ro after the blocking can interfere the execution of lower priority tasks one more times than the original RMA test. The schedulability of the task set of Example 1 is analyzed with this approach as follows. The task rb is considered as the two subtasks r and r2 ....
R. Rajkumar, L. Sha, and J.P. Lehoczky, "Real-time synchronization protocols for multiprocessors," in Proc. Real-Time Systems Symposium, pp. 25969, Dec. 1988.
....scheduling algorithms is that only independent tasks that do not synchronize or share resources have been considered. In contrast, tasks in real systems usually are not independent. Synchronization entails additional overhead, which must be taken into account when determining system feasibility [2, 6, 18, 19, 20, 21]. Unfortunately, prior work on real time synchronization has been directed at uniprocessor systems, or systems implemented using non fair scheduling algorithms (or both) and thus cannot be directly applied in fair scheduled multiprocessor systems. Indeed, fair scheduled uniprocessor systems ....
....systems. We intend to examine these lock based techniques in more detail in future work. In real time systems in which locks are used, priority inversions must be dealt with. A priority inversion occurs when a task is blocked by a task of lower priority. Inheritance and ceiling schemes [6, 18, 19, 20, 21] limit the duration of priority inversions by temporarily boosting a lock holding task s priority when it blocks any higher priority task. In fair scheduled systems, such schemes cannot be applied directly because they disrupt allocation rates. As a result, blocking times in fair scheduled systems ....
R. Rajkumar, L. Sha, and J. Lehoczky. Real-time synchronization protocols for multiprocessors. In Proceedings of the Ninth IEEE Real-Time Systems Symposium, pages 259--269. 1988.
.... research on QoS for network infrastructure has focused largely on policies for allocating bandwidth on a perconnection basis [43] Likewise, research on real time operating systems has focused largely on avoiding priority inversions and non determinism in synchronization and scheduling mechanisms [44]. In contrast, the programming model for developers of OO middleware focuses on invoking remote operations on distributed objects. Determining how to map the results from the network and OS layers to OO middleware is the main focus of the TAO research project. There are several commercial ....
R. Rajkumar, L. Sha, and J. P. Lehoczky, "Real-Time Synchronization Protocols for Multiprocessors," in Proceedings of the Real-Time Systems Symposium, (Huntsville, Alabama), December 1988.
.... processors while the communication between subtasks can be modeled as a task by itself [24] Rajkumar et al. describe a method to schedule tasks statically assigned to processors in a shared memory system but they allow proxy execution of critical sections to handle requests for remote resources [19]. These approaches limit the general model of shared resources while we make no assumptions about the location of resources or task interdependencies. We focus on a protocol to acquire resources in a distributed environment, bound the message overhead and provide support to prevent priority ....
....e.g. tasks with shared resources are required to be scheduled on the same processor. Rajkumar et al. describe a method to schedule tasks statically assigned to processors in a shared memory system but they allow proxy execution of critical sections to handle requests for remote resources [19]. For remote resources in multi processors, real time scheduling approaches often utilize heuristic methods since the general problem of distributed scheduling is NP complete [21] Heuristic algorithms have also been proposed for distributed systems to select nodes for tasks when they become ready ....
R. Rajkumar, L. Sha, and J. Lehoczky. Real-time synchronization protocols for multiprocessors. In IEEE Real-Time Systems Symposium, pages 259--269, Dec. 1988.
No context found.
Rajkumar, R., Sha, L., and Lehoczky, J.P. "Real-Time Synchronization Protocols for Multiprocessors". IEEE Real-Time Systems Symposium (December 1988).
No context found.
R. Rajkumar and L. Sha and J. P. Lehoczky, "Real-Time Synchronization Protocols for Multiprocessors ", In IEEE Real-Time Systems Symposium, pages 259-269, 1988.
No context found.
R. Rajkumar, L. Sha, and J.P. Lehoczky. Real-Time Synchronization Protocols for Multiprocessors. In Proceedings of Real-time Systems Symposium, December 1988.
No context found.
R. Rajkumar, L. Sha, and J. P. Lehoczky. Realtime synchronization protocols for multiprocessors. In Proceedings of IEEE 9th Real-Time Systems Symposium, pages 259--269, December 1988.
No context found.
R. Rajkumar, L. Sha, and J. Lehoczhy, "RealTime Synchronization Protocols for Multiprocessors, " in Proc. of the Real-Time System Symposium, pp. 259--272, 1988.
No context found.
R. Rajkumar, L. Sha, and J. P. Lehoczky, "Real-time synchronization protocols for multiprocessors," in IEEE Real-Time Systems Symposium, Dec. 1988, pp. 259--269.
No context found.
R. Rajkumar, L. Sha, and J. P. Lehoczky, "Realtime synchronization protocols for multiprocessors," Real Time Systems Symposium, pp. 259--269, December 1988.
No context found.
R. Rajkumar, L. Sha, J.P. Lehoczky. Real-time synchronization protocols for multiprocessors. In Proc. 9th IEEE Real-Time Systems Symposium, pages 259-269, 1988.
No context found.
R. Rajkumar, L. Sha, and J.P. Lehoczky. Real-Time Synchronization Protocols for Multiprocessors. In Proceedings of Real-time Systems Symposium, December 1988.
No context found.
R. Rajkumar, L. Sha, and J.P. Lehoczky. "Real-Time Synchronization Protocols for Multiprocessors". IEEE Real-Time Systems Symposium, December 1988.
No context found.
R. Rajkumar, L. Sha, and J. Lehoczky. Real-time synchronization protocols for multiprocessors. In IEEE Real-Time Systems Symposium, pages 259-269, December 1988.
No context found.
Rajkumar, R., Sha, L., and Lehoczky, J. P., "Real-time synchronization protocols for multiprocessors," Real Time Systems Symposium, pp. 259--269, December 1988.
First 50 documents Next 50
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC