D. R. Cheriton and W. Zwaenepoel, "The distributed V Kernel and its performance on diskless workstations," in Proceedings of the Ninth Symposium on Operating Systems Principles, pp. 129--140, ACM, October 1983.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....in porting V on a multiprocessor machine, connecting it to wide area networks and implementing reliability issues such as atomic transactions and replication. 48 Contact: David R. Cheriton or Tony Mason, Computer Science Department, Stanford University, Stanford CA 94305. References: 318] [319], 320] 321] 322] 323] 324] 325] 326] 327] 328] 329] 330] 331] 332] 333] 146] 334] 29] 335] 336] 2.51 VAXcluster Main Goal The Digital Equipment Corporation has been developed the VAXcluster product as a set of closely coupled DEC VAX computers providing a ....

D.R. Cheriton and W. Zwaenepoel, "The Distributed V Kernel and its Performance for Diskless Workstations ", In Proc. 9th Symp. on Operating Systems Principles, pages 129--140, ACM, October 1983.

....trans( operations use the put port. The kernel then compares the message destination port with its internal tables to find the correct server. This mechanism was developed as a security mechanism to prevent arbitrary applications from receiving messages on the server s chosen port. 3. 4 V V [17] was developed from scratch as a distributed kernel meant to be used for a network of diskless workstations. It has a message based communication model using a synchronous send reply protocol. With its emphasis on the distributed environment, the IPC mechanism is a fundamental component of the ....

D. Cheriton and W. Zwaenepoel, "The distributed V kernel and its performance for diskless workstations," in Proceedings of the 9th ACM Symposium on Operating Systems Principles, pp. 129--140, October 1983.

....server. The local name server caches the remote address for possible future reuse and sends the client the address of the bridge on the local network in the NAME LOOKUP REPLY. Again the client and server are completely unaware of the bridges between their networks. 4.2. The V System The V System [9] is a distributed operating system running on a collection of processors connected by an Ethernet. The processors are divided into two types: workstations and server machines. The workstations are like processors in the Cambridge Processor Bank, except that workstations have owners that have ....

D. R. Cheriton and W. Zwaenepoel, "The Distributed V Kernel and its Performance for Diskless Workstations," Proc. of the 9th ACM Symp. on Operating Systems Principles, pp. 128-140, Bretton Woods, NH, October 1983.

....I O intensive applications, such as multimedia, parallel file systems, and supercomputing on clusters of workstations, among others, demand elimination of this bottleneck. Several previous works have proposed improving data passing efficiency by making I O operations have move [22] or share [5, 2, 9] semantics. Move semantics avoids data copying by using typically much cheaper virtual memory (VM) manipulations. On output with move semantics, the system removes the region containing the application data from the application address space. The application buffer becomes the system output ....

....could cause deadlock if the faulted process is itself a (possibly userlevel) pager or is invoked by a pager. In our implementation, input disabled pageout eliminates wiring costs safely and cheaply. Programmed I O may have the advantage of simplifying the implementation of early demultiplexing [5, 2]. There have been proposals to reduce the penalty of copying by integrating it with other data touching operations, such as TCP checksumming [6] Integration of checksumming on input has semantic implications: If checksumming is integrated with the copy from device or system buffer to ....

D. Cheriton and W. Zwaenepoel. "The Distributed V Kernel and its Performance for Diskless Workstations", in Proc. 9th SOSP, ACM, Oct. 1983, pp. 129-140.

....modelled in the earlier work. The method is several orders of magnitude faster than Petri net analysis even on small examples. The accuracy obtained is generally better than 10 . 1 Introduction In concurrent software written for multi tasking, multi processing environments such as Ada [DOD83] V [CHER83], or XMS [GAMM84] we often find a sequence of tasks operating as a pipeline on a stream of data. For example, a radar tracking system might have tasks for filtering, detection, track correlation, track update, display and storage updating for the track. It may be a simple sequence of tasks as ....

D. R. Cheriton, W. Zwaenpol, "The distributed V-kernel and its performance for diskless workstations", Proc. 9th Symp on Operating Systems Principles, ACM Operating Systems Review v17 n 5, pp 129-140, Oct 1983.

....be realized here and there, but there does not seem to be any way to achieve significant performance gains without eliminating language features. Work by other researchers tends to confirm the hypothesis that data transmission times do not dominate the cost of practical message passing systems [9, 18, 21, 30]. High level semantic functions such as addressing, dispatching, bookkeeping, testing, and error handling are at least as significant, and often more so. One millisecond appears to be a nearly universal lower bound on round trip communication times with mid 1980s microprocessor based ....

D. R. Cheriton and W. Zwaenepoel, "The Distributed V Kernel and its Performance for Diskless Workstations," Proceedings of the Ninth ACM Symposium on Operating Systems Principles, 10-13 October 1983, pp. 129-140. In ACM SIGOPS Operating Systems Review 17:5.

.... Several distributed schedulers have been designed to balance the goal of improving performance with that of preserving predictability of service, including Condor [10] Sprite [4, 12] and the Process Server [7] In addition, several systems, including Butler [3, 11] NEST [1] and V SYSTEM [2], have provided mechanisms to support remote execution of processes in the WDS environment, though they have not provided policies to control these mechanisms. All these systems share a common conservative approach to preserving predictability of service: Foreign processes (those executing at a ....

D. R. Cheriton and W. Zwaenepoel, "The Distributed V Kernel and its Performance for Diskless Workstations," Proceedings of the Ninth ACM Symposium on Operating Systems Principles, pp. 128-139 (10-13 October 1983) In ACM Operating Systems Review 17:5.

No context found.

D.R. Cheriton and W. Zwaenepoel, "The Distributed V Kernel and its Performance for Diskless Workstations", In Proceedings of the 9th Symposium on Operating System Principles, ACM, 1983.

No context found.

D. R. Cheriton and W. Zwaenepoel, `The distributed V kernel and its performance for diskless workstations', Proceedings of the 9th ACM Symposium on Operating Systems Principles, October 1983, pp. 129--140.

....run on a collection of SUN 2 workstations connected by a 10 megabit Ethernet. During the experiments the machines are exclusively used by the programs involved in the experiment. The machines are running the V System, an experimental message based operating system developed at Stanford University [3]. Interprocess communication is by means of messages and is transparent (i.e. independent of the location in the network of the communicating processes) 4 Measurements We have compiled and measured several programs using the dynamic and the combined evaluators, both sequentially and in parallel ....

D. R. Cheriton and W. Zwaenepoel, "The Distributed V Kernel and its Performance on Diskless Workstations ", Proceedings of the Ninth ACM Symposium on Operating System Principles, pp. 128-140 (1983).

No context found.

D. R. Cheriton and W. Zwaenepoel, "The distributed V Kernel and its performance on diskless workstations," in Proceedings of the Ninth Symposium on Operating Systems Principles, pp. 129--140, ACM, October 1983.

No context found.

D. R. Cheriton and W. Zwaenpoel: "The Distributed V kernel and its performance for diskless workstations", Proc. 9th Symp. on Operating System Principles, pp. 129-40, Oct. 1983

No context found.

D. R. Cheriton and W. Zwaenepoel, `The distributed V kernel and its performance for diskless workstations', Proc. 9th ACM Symposium on Operating System Principles, Bretton Woods, New Hampshire, 10--13 October 1983, pp. 128--140.

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