Matthew J. Zelesko and David R. Cheriton. Specializing object-oriented RPC for functionality and performance. In Proceedings of the 16th IEEE International Conference on Distributed Computing Systems, pages 175--187. IEEE Computer Society Press, May 1996. 18  Home/Search   Document Details and Download   Summary   Related Articles   Check

....transfer of objects. Similar to RPC [37, 38] servers contain implementations of procedures. In objectoriented concepts these procedures are called methods. A client calls these methods and waits until they are finished. There already have been objectoriented approaches with RPC in advance of RMI [42]. Because RMI allows only unidirectional interaction of two nodes it has to establish two communication channels where each node contains a client and a server. Before the network transports data structures, RMI marshals them into byte streams. After the receiver obtained all data the byte stream ....

Zelesko, M. J., and Cheriton, D. R. Specializing Object-Oriented RPC for Functionality and Performance. In Proceedings of the 1996 International Conference on Distributed Computing Systems (Hong Kong, May 1996), pp. 175--187.

....our work, they build a set of components that can be assembled to provide various flavors of DCOM. Another closely related work is on TAO ORB [16] by Schmidt et al. They abstract ORB functionality as design patterns and use them to customize their ORB for hard real time features. The OORPC system [25], a customizable RPC implementation, also shares some of our ideas and demonstrates the benefits of RPC customization. We have also used Quarterware based implementations for CORBA extensions for real time and fault tolerance [19] and object group management for SUN Netra Proxy servers [18] The ....

Matthew J. Zelesko and David R. Cheriton. Specializing object-oriented RPC for functionality and performance. In Proceedings of the 16th International Conference on Distributed Computing Systems, pages 175--187, 1996.

....blocking synchronization. ffl First, NBS allows synchronized code to be executed anywhere. Non blocking synchronization allows synchronized code to be executed in an interrupt or (asynchronous) signal handler without danger of deadlock. For instance, an asynchronous RPC handler (as described in [103]) can directly store a string into a synchronized data structure such as a hash table even though it may be interrupting another thread updating the same table. With locking, the 1.1. COORDINATION AMONG COOPERATING TASKS 7 signal handler could deadlock with this other thread. ffl Second, ....

.... are the only kernel supported form of notification, allowing a simple, efficient kernel implementation compared to more complex kernel message primitives, such as those used in V [23] Class libraries implement higher level communication like RPC in terms of signals and shared memory regions [103]. Non blocking synchronization allows efficient library implementation without the overhead of disabling and enabling signals as part of access and without needing to carefully restrict the code executed by signal handlers. Second, we simplified the kernel and allowed specialization of these ....

[Article contains additional citation context not shown here]

Zelesko, Matt and David R. Cheriton. "Specializing Object Oriented RPC for Functionality and Performance." In Proceedings 16th IEEE International Conference on Distributed Computing Systems. IEEE Computer Society Press, May 27-30 1996.

....of the Cache Kernel [7] operating system kernel and supporting libraries for several reasons. First, non blocking synchronization allows synchronized code to be executed in an (asynchronous) signal handler without danger of deadlock. For instance, an asynchronous RPC handler (as described in [25]) can directly store a string into a synchronized data structure such as a hash table even though it may be interrupting another thread updating the same table. With locking, the signal handler could deadlock with this other thread. Second, non blocking synchronization minimizes interference ....

.... are the only kernel supported form of notification, allowing a simple, efficient kernel implementation compared to more complex kernel message primitives, such as those used in V [6] Class libraries implement higher level communication like RPC in terms of signals and shared memory regions [25]. Non blocking synchronization allows efficient library implementation without the overhead of disabling and enabling signals as part of access and without needing to carefully restrict the code executed by signal handlers. Second, we simplified the kernel and allows specialization of these ....

[Article contains additional citation context not shown here]

M. Zelesko and D. R. Cheriton, Specializing Object Oriented RPC for Functionality and Performance, Proceedings 16th IEEE International Conference on DistributedComputingSystems, IEEE Computer Society Press, May 27-30, 1996.

No context found.

Matthew J. Zelesko and David R. Cheriton. Specializing object-oriented RPC for functionality and performance. In Proceedings of the 16th IEEE International Conference on Distributed Computing Systems, pages 175--187. IEEE Computer Society Press, May 1996. 18

No context found.

Matthew J. Zelesko and David R. Cheriton. Specializing object-oriented RPC for functionality and performance. In Proceedings of the 16th IEEE International Conference on Distributed Computing Systems, pages 175--187. IEEE Computer Society Press, May 1996. 18

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