The Advanced Network Systems Architecture, A.J.Herbert 1989, "Distributed Systems" ed. Sape Mullender, ISBN 0-201-41660-3.  Home/Search   Document Not in Database   Summary   ACM   TOC   Related Articles   Check

....filtering performed by filtering indirectors. We are not aware of such filters being used to defer packet processing, nor do they appear to have been used to filter dynamically tagged messages. Use of registers to speed interprocess communication was heavily used in the V Distributed System [4], and appears to have been independently proposed by Karger [16] in connection with the SCAP system [17, 15] The SCAP design gains particular advantage if the trust relationship between caller and callee is known to both parties. Liedtke et al. 22] have considered selected denial of service ....

D. Cheriton. The v distributed system. (3), Mar. 1988.

....of delegation issues [VAB91] Current delegation systems have various tradeoffs in scalability, revocation and processing requirement; the method we present minimizes processing and enables frequent revocation. Other systems use message ports for communication, including Mach [Ras86] and V [Che88]. Anderson and Rangan have developed an authentication interface based on message ports, though it is incompatible with previous interfaces [AR87] None of these systems are able to benefit from object orientation to maximize code reuse and provide polymorphism. Furthermore, it is not clear how to ....

David R. Cheriton. The V Distributed System. Communications of the ACM, 31(3):314--333, March 1988.

....to this problem: Many monolithic kernels such as BSD, Linux, and Windows NT implement the process model, where each thread has its own kernel stack; when the thread is not running, most of its state is implicitly encoded in this stack. In the interrupt model, which has been used e.g. in V [8], QNX [21] and Fluke [16] the kernel uses only one kernel stack per processor. Threads are required to record their state in an explicit kernel object, a continuation, before blocking. Traditionally, L4 implementations have used the process model, i.e. per thread kernel stacks. However, as ....

David R. Cheriton. The V distributed system. Communications of the ACM, 31(3):314--333, 1988. ISSN 0001-0782.

....system. Its system architecture is orga nized around a processor pool . All users have equal access to all pool processors. Pool processors are dynamically allocated to processes as needed. Amoeba achieves some load balancing by assigning the most desirable processor to a process. The V System [19] uses a workstation model similar to Sprite. It uses process migration to execute new tasks on lightly loaded workstations. There are many other systems that attempt to make use of idle computing power through a remote execution facility. The Benevolent Bandit Laboratory (BBL) 26] runs ....

....although the set of idle machines changes over time, the total number of idle machines stays relatively con stant. Our objective is to use the idle workstations to run additional jobs. There have been several systems that attempt to make use of idle workstations to execute sequential programs [19, 44, 58]. In systems using idle workstations, the additional computation is suspended when primary user activity is detected to avoid perfor mance degradation for primary users. The additional computation is resumed when primary user activity ends and the workstation is idle. Since the workstations are ....

D. R. Cheriton. The V Distributed System. Comm. of the ACM, 31(3):314-333, March 1988.

....the fast path is abandoned more often by lock conflicts on a uniprocessor. Even with just one caller thread, the interrupt code can require a lock held by the caller or server thread. It is plausible that better uniprocessor throughput could be achieved by an RPC design, like Amoeba s [9] V s [3], or Sprite s [6] that streamed a large argument or result from a single call in multiple packets, rather than depended on multiple threads transferring just a packet s worth of data per call. The streaming strategy requires fewer thread to thread context switches. 6. OTHER SYSTEMS A sure ....

....ACM Transactions on Computer Systems, Vol. 8, No. 1, February 1990. 16 M. Schroeder and M. Burrows Table XII. Performance of Remote RPC in Other Systems System Machine Processor MIPs ms call Mbits s Cedar [2] Dorado Custom I x 4 1.1 2.0 Amoeba [9] Tadpole M68020 I x 1.5 1.4 5. 3 V [3] Sun 3 75 M68020 I x 2 2.5 4.4 Sprite [6] Sun 3 75 M68020 I x 2 2.8 5.6 Amoeba UNIX 4 [9] Sun 3 50 M68020 I x 1.5 7.0 1.8 Firefly FF MicroVAX II I x I 4.8 2.5 Firefly FF MicroVAX II 5 x i 2.7 4.6 It is clear from the literature that developers of distributed systems are learning how to get ....

CHERITON, D.R. The V distributed system. Commun. ACM 31, 3 (Mar. 1988), 314-333.

....many different techniques have been demonstrated. In this section, we discuss some of the research directions, and how they relate to this work. Virtual memory Some approaches use virtual memory primitives to optimize the transfer of large messages. Peregrine [23] which is based on the V system [6], uses page remapping to efficiently move the call arguments and results between the client s and server s address spaces. The DEC Firefly RPC facility [32] uses a global buffer pool, which resides in memory shared among all user address spaces, whereas in the fbuf scheme presented by Druschel and ....

David Cheriton. The V distributed system. Communications of the ACM, 31(3):314--333, 1988.

....for future work. 2 Related Work Many research operating systems have been developed that implemented process migration mechanisms, with a focus on using migration for load balancing. These sys tems include Accent [31] Amoeba [25] Charlotte [6] Chorus [33] MOSIX [7] Sprite [12] and V [11]. These operating systems provided a single system image across a cluster of machines and providing migration throughout the cluster through careful kernel design to provide a global namespace and location transparent execution. Process state such as IPC, open files, and system calls in some cases ....

D. Cheriton, The V Distributed System, Communications of the ACM, 31(3):314-333, March 1988.

....size partitions [72] Systems such as Charm [91] the Parform [21] PVM Hence [96] and others [29, 42, 44, 97] support parallel computing on a network of workstations. In these systems, the set of machines on which the program runs is chosen statically by the user. Distributed operating systems [30, 82, 98, 99] and other systems [32, 40, 68, 74, 79, 110] provide transparent process placement and (in some cases) migration, but these systems are geared towards large serial programs or coarse grain distributed programs. A system that does provide adaptive parallelism is Piranha [23, 46, 62] The creators ....

David R. Cheriton. The V distributed system. Communications of the ACM, 31(3):314--333, March 1988.

....175 198. 176 B. N Bershad et al. General Terms: Design, Performance, Measurement Additional Key Words and Phrases: Modularity, remote procedure call, small kernel operating system 1. INTRODUCTION Efficient interprocess communication is central to the design of contemporary operating systems [16, 23]. An efficient communication facility encourages system decomposition across address space boundaries. Decomposed systems have several advantages over more monolithic ones, including failure isola tion (address space boundaries prevent a fault in one module from leaking into another) ....

....of data copying and marshalling that is incurred when parameters are passed between address spaces. On the Firefly, each word of data passed in the shared memory buffers adds about one additional sec of latency Because the amount of data passed in most cross address space procedure calls is small [8, 16, 17, 24], average call performance is most influenced by the components shown in Table II, and not the cost of data transfer. 4.3 Call Latency and Throughput The figures in Table II are independent of client and server load, provided there is no need for processor reallocation. Two other important ....

CERON, D.R. The V distributed system. Commun. ACM. 31, 3 (Mar. 1988), 314-333.

No context found.

The Advanced Network Systems Architecture, A.J.Herbert 1989, "Distributed Systems" ed. Sape Mullender, ISBN 0-201-41660-3.

No context found.

D. R. Cheriton. The V Distributed System. Communications of the ACM, 31(3), March 1988.

No context found.

D. Cheriton. The v distributed system. Communications of the ACM, 31(3), March 1989.

No context found.

Cheriton, D. R. The V Distributed System. In Communications of the ACM, pp. 314--333, March 1988.

No context found.

Cheriton, D. R. The V Distributed System. In Communications of the ACM, pp. 314--333, March 1988.

No context found.

D. Cheriton. The v distributed system. Communications of the ACM, 31(3):314--333, March 1988.

No context found.

David R. Cheriton. The V distributed system. Communications of the ACM, 31(3), March 1988.

No context found.

D. Cheriton. The V Distributed System. Communications of the ACM, 31(3):314--333, Mar 1988.

No context found.

CHERITON, D. The v distributed system. Communications of the ACM (CACM) 31, 3 (March 1988).

No context found.

D. Cheriton. The V distributed system. Communications of the ACM, 31(3):314--333, Mar 1988.

No context found.

David R. Cheriton. The V Distributed System. Communications of the ACM, 31(3):314-333, March 1988. 71

No context found.

David Cheriton. The V distributed system. Communications of the ACM, 31(3):314--333, March 1988.

No context found.

David Cheriton. The V Distributed System. Communications of the ACM, 31(3), March 1988, pp.314-333.

No context found.

David R. Cheriton. The V distributed system. Communications of the ACM, 31(3):314--333, March 1988.

No context found.

D. R. Cheriton. The V Distributed System. Communications of the ACM, 31(3):314-333, March 1988.

No context found.

David R. Cheriton. The V Distributed System. In Communications of the ACM, pages 314--333, March 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