A. D. Birrel and B. J. Neslon. Implementing Remote Procedure Calls. ACM Trans. on Computer Systems, 2:39--59, 1984.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....not make it inefficient for transactions, while maintaining the things that make it good for data streaming. T TCP employs two optimizations to deal with these inefficiencies. These two optimizations essentially incorporate two techniques from the implementation of RPC by Birrell and 25 Nelson [5] into TCP. The first optimization, known as TCP Accelerated Open (TAO) eliminates the need for the three way handshake protocol at the opening phase of communication for most instances. Figure 2 2 illustrates the TAO mechanism. This optimization is accomplished by using a dual monotonic number ....

Andrew Birrell and Bruce Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1), 1984.

....appropriate abstractions for distributed application programmers and developing run time systems that support these abstractions. Examples of successful paradigms include distributed object oriented computing (e.g. CORBA [129] group communication (e.g. ISIS [16] remote procedure calls (RPC [18]) and distributed shared memory (e.g. MUNIN [24] These paradigms presented convenient new entities of which the programmer s world is composed (e.g. objects and process groups) and implemented mechanisms for their interaction. Current paradigms for distributed computing, however, share in ....

A.D. Birrel and B.J. Nelson, "Implementing Remote Procedure Calls," ACM Transactions on Computer Systems , Vol 2, No. 1, February 1984.

....tunneled, where traOEc both to and from the mobile host goes through the home agent. 3.5 Service Centric networking Classic network applications work by opening a socket from a client to a server. The application has to deal with the protocol on this socket itself. Remote Procedure Call(RPC)[5], Remote Method Invocation(RMI) 47, 38] and Simple Object Access Protocol (SOAP) 44] are software modules which enables the possibility of remotely calling procedures in a server. Using these technologies, a server application may export a method interface which clients can use. Middleware in the ....

Andrew D. Birrel and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):3959, February 1984.

....mean that the two styles of RPC communication are not necessarily equally applicable. The following sections examine both message passing and thread tunnelling in the context of the Rcane architecture. 5.6.1. 1 Message Passing RPC between two separated computers, as presented originally in [Birrell84] requires message passing. The sequence of events which occur during a message passing RPC is as follows: In fact the argument in [Lauer79] is expressed in terms of monitors ( Hoare73] rather than threadtunnelling however, a similar duality can be drawn between a thread tunnelling system ....

....any results or exceptions into a reply bu#er and transfers it back to the client. 3. The client thread is unblocked, unmarshals the results, and returns to its caller or raises an exception as appropriate. Many traditional RPC systems both inter and intra machine use message passing [Birrell84, Birrell93, Roscoe95] 5.6.1.2 Thread Tunnelling The time taken to reschedule to the server, and then back to the client, can add significant latency to the time taken for a message passing RPC to complete. Furthermore, the server may not be eligible to receive CPU time at the point of the ....

Andrew D. Birrell and Bruce Jay Nelson. Implementing Remote Procedure Calls. ACM Transactions on Computer Systems, 2(1):39--59, February

....vision for next generation programming systems supporting future real time deeply embedded distributed sensor network applications. 4 Related Work Classical distributed programming paradigms and middleware such as CORBA [27] group communication (e.g. ISIS [5] remote procedure calls (RPC [6]) and distributed shared memory (e.g. MUNIN [9] share in common the fact that their programming abstractions exist in a logical space that does not represent or interact with objects and activities in the physical world. Their main goal is to abstract distributed communication rather than ....

A. Birrel and B. Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1), February 1984.

....embedded in a physical environment with which they tightly interact. It is this last difference that motivates the programming paradigm described in this paper. Classical distributed programming paradigms and middleware such as CORBA [33] group communication [8] remote procedure calls [4], and distributed shared memory [6, 30] share in common the fact that their programming abstractions exist in a logical space that does not represent or interact with objects and activities in the physical world. This is expected since their main goal is to abstract distributed communication ....

A. Birrel and B. Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1), February 1984.

....an area for future work. In contrast, some popular applications do not maintain any stateful relationships between application end points. Such so called stateless network applications often use non connection oriented transfer protocols (e.g. UDP) and asynchronous remote procedure calls (RPC) [14]. Transport 26 protocol alone does not necessarily indicate whether an application is session oriented or not, however. We present Migrate s support for non connected UDP sockets used in a session oriented fashion in Section 4.4) These communication paradigms have little to gain from ....

....explicitly communicate with remote application end points require a far greater level of support. In some instances, applications can defer communication yet continue to provide a user with some level of functionality. In particular, applications based on the Remote Procedure Call (RPC) model [14], where each communication is a request reply exchange, have been successfully adapted for disconnected operation using the Rover toolkit [51] which queues RPC for later delivery. In addition to queuing RPCs, the Rover toolkit can emulate Coda s hoarding process; instead of copying or relocating ....

[Article contains additional citation context not shown here]

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, February 1984.

....unacceptable for a multi server system. Moreover, programmers may be tempted to base their design decisions on communication cost; as Birrell and Nelson have pointed out, RPC performance] seems important, lest communication become so expensive that application designers strenuously avoid it [4]. In this paper, we present an approach that improves stub code performance by specialization for the underlying platform. We exploit specific knowledge of the microkernel, the hardware architecture, and the compiler that are being used, as well as certain characteristics of the application ....

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems (TOCS), 2(1):39--59, 1984.

....to the client using Globus I O for subsequent parameter transfer, etc. Subsequent remote executable communication with the client will use this port. 4 Related Work The concept of Remote Procedure Call (RPC) has been widely used in distributed computing and distributed systems for many years [11]. It provides an elegant and simple abstraction that allows distributed components to communicate with well de ned semantics. RPC implementations face a number of dicult issues, including the de nition of appropriate Application Programming Interfaces (APIs) wire protocols, and Interface ....

A. Birrel and G. Nelson. Implementing Remote Procedure Calls. ACM Transactions on Computer Systems (TOCS), 2(1):39-59, 1984.

....of these applications. Furthermore, the loss of flexibility (inherent to the use of any high level abstraction) and the performance overhead, are negligible with respect to the simplicity gained by using TPS. 1. Introduction Remote Procedure Call (RPC) was first proposed by Birrel and Nelson [1] as a simple abstraction that conceals interactions between remote components beneath traditional procedural interfaces. Partly because of its simplicity and the very facts that it preserves object encapsulation and ensures type safety, and partly because RPC s overhead was very acceptable over ....

Andrew D. Birrell and Bruce Jay Nelson. Implementing Remote Procedure Calls. In Proceedings of the 9th ACM Symposium on Operating Systems Principles (SOSP'83). October 1983.

....performance. The faster I O devices now available benefit in terms of performance and or 31 functionality by appropriately tailored OS support for communication. Much work has been done on IPC over the past 15 years, either to enrich the primitives, or purely for performance reasons e.g. [Birrell84, Karger89, Bershad90, Schroeder90, Liedtke93, Johnson93]. This work is relevant to the provision of QoS isolated traffic streams, because any system with protection boundaries on the data path needs to ensure that IPC does not become a bottleneck, as argued in [Hsieh93] 2.3.1 Lightweight IPC The early work on RPC [Birrell84, Schroeder90, Johnson93] ....

....Schroeder90, Liedtke93, Johnson93] This work is relevant to the provision of QoS isolated traffic streams, because any system with protection boundaries on the data path needs to ensure that IPC does not become a bottleneck, as argued in [Hsieh93] 2.3. 1 Lightweight IPC The early work on RPC [Birrell84, Schroeder90, Johnson93] focused on making the non local interactions appear to the programmer as simple procedure calls. They are all message based, since ultimately messages must cross the network. Research into local only RPC systems was prompted by the growth in popularity of microkernels such as Mach 3.0, Amoeba, ....

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, February

....range of memory a process is allowed to access. The permission entries in the TLB can also be used to restrict memory accesses, provided that only the kernel is allowed to manipulate TLB entries. Calls between protected domains are handled by the kernel through a remote procedure call mechanism, [3, 4]. 3.3 Software Fault Isolation Software fault isolation, or sandboxing [28] was introduced to avoid the high costs incurred by placing untrusted code in its own address space and accessing it via remote procedure calls. The untrusted code is modified to include address checks before each ....

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, February 1984.

....Jini [1] and ONE [20] technologies, and SOAP UDDI [4, 8] While the term service oriented architecture is to our knowledge a somewhat recent one, related technologies have been the topic of extensive research since at least the early 1970 s. One such technology is the remote procedure call (RPC) [3], which allows applications to invoke remote functions as if they were local functions. This feature isolates the developer from di#erences in operating systems, implementation languages, and network protocols used between the various hardware and software components that comprise the ....

Andrew D. Birrell and Bruce J. Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, February 1984.

....the actual response time to bandwidth. For example, if a 65KB packet is sent cross country on a 1.2Gb s channel while simultaneously sending the same data over a 10Mb s Ethernet, the high speed transmission will complete before the Ethernet does, despite the high latency. Remote Procedure Call [Birrell Nelson] or RPC) has received considerable attention in the effort to reduce latency. A new version of RPC for the Firefly [Thacker, et al. 88] multiprocessor which breaks down the time used for each of 15 different steps in an RPC operation can be found in [Schroeder Burrows 89] The objective was ....

Andrew D. Birrell & Bruce Nelson. Implementing Remote Procedure Calls. ACM Transactions on Computer Systems, February 1984, pp. 39-59.

....access data is huge) and will have so much internal processing capability that HTMT programming models may wish to view them as some sort of servers with the programs running in the SPELLs as clients. 5. 3 Remote method invocation Remote Procedure Call (RPC) and Remote Method Invocation (RMI) [10] allow programs to call procedures located on other machines. When a process on machine A calls a procedure on machine B, the calling process on A is suspended, and the execution of the called procedure takes place on B. Information is transported through the stub (files that are created by ....

Birrel, A., B.J. Nelson, "Implementing Remote Procedure Calls," ACM Transactions on Computer Systems, vol. 2, no. 1, pp. 39-59, Feb. 1984.

....to allow this form of fragmented access. A client interface fragment can then be used as a named reference and is responsible for maintaining reference validity and for any transparent or independent locality that the model may support. Fragmented objects are similar to concepts used in RPC [20, 22], CORBA [138] DCOM [27] and Java RMI [167] The fragmented object model di#ers from these in that it provides support for connective objects and group interfaces that enables multiple levels of fragmentation, while the other systems provide only one level of fragmentation: separating client ....

....used to bootstrap a client into an ORB system by enabling it to specify attribute based resolution requests with full knowledge of the attributes used within the naming system. 6.6 Communication Models The DISCWorld ORB system supports two models of communication. The first model is an RPC style [20, 22] model involving a fragmented object model. In this model, the client interface fragment acts as a client side proxy for the server object. The proxy is able to CHAPTER 6. THE DISCWORLD ORB SYSTEM 166 transparently invoke methods, both synchronously and asynchronously, on the remote server object. ....

Andrew D. Birrell and Bruce Jay Nelson. Implementing Remote Procedure Calls. ACM Transactions on Computer Systems, Volume 2, Number 1, pages 39--59, February 1984.

....is currently visiting the Swiss Federal Institute of Technology in Lausanne. 1. INTRODUCTION 1. 1 RPC et al. One of the most popular styles of distributed programming relies on extending the notion of invocation to a distributed context, i.e. o#ering some form of remote procedure call (RPC) [7]. The integration of this distributed interaction style with an object oriented programming language has been thoroughly studied, e.g. 41, 11, 42, 10] More recently, Java [29] has introduced its own variant, the remote method invocation (RMI) 61] through a precompiler approach. By using the ....

....easily be banned, by not considering a subscription as part of the state of a subscribed obvent: once published, the copies of an obvent would have to reissue any subscriptions performed by their master copy . 5. 4 RPC and Publish Subscribe Originally introduced as remote procedure call (RPC) [7], remote invocations have been quickly applied to objectoriented languages, leveraging some form of remotely accessible entity, e.g. guardians in Argus [41] its follow up CLU [42] network objects in Modula 3 [11] and Obliq [10] every object is potentially a network object) and of course ....

A. Birrel and B. Nelson. Implementing Remote Procedure Calls. ACM Transactions on Computer Systems, 2(1):39--59, Feb. 1984.

....supported by the United States National Security Agency University Research Program and by ZWorld, Inc. and the State of California under the UC MICRO Program. Many systems contain mechanisms for performing distributed communication. They typically provide some kind of Remote Procedure Call (RPC) [1]. For example, DCOM[2] Java RMI[3] ILU[4] and CORBA[5] are forms of object request brokers that allow remote methods to be invoked through remote objects. Direct support for interoperability of concurrency control mechanisms under these systems is limited. For example, CORBA s Concurrency ....

Andrew Birrell and Bruce Nelson. Implementing remote procedure calls. ACM Trans. Computer Systems, 2(1):39--59, February 1984.

.... Parnas [1972] ffl specification languages (e.g. Mosses [1999] ffl programming languages, both procedural (e.g. UCSD Pascal, Ada) and object oriented (e.g. Simula, Smalltalk, Eiffel) ffl object oriented analysis and design (e.g. Deutsch [1989] ffl remote procedure call systems (e.g. Birrel and Nelson [1984]) ffl component infrastructures (e.g. Szyperski [1997] and ffl object request brokers (e.g. Buschmann et al. 1996] OMG [1999] In the last few years, the interface concept has come of age. It has become clear that interfaces are an important organising idea at a number of levels of ....

Birrel and Nelson [1984] A D Birrel and B J Nelson. Implementing Remote Procedure Calls. ACM Transactions on Computer Systems, 2(1):39--59, 1984.

....distributed I O systems make use of a centralized server to connect jobs with data. The canonical example is of course the Network File System (NFS) 29] An analogue in grid computing is the Condor [20] remote system call [19] facility, in which each running job performs remote procedure calls [9] back to the orginating computer. Both of these central server models have limited scalability, because the number of clients is limited by the aggregate bandwidth provided by the central server. The performance of individual clients may also be limited by the bandwidth or the latency of the ....

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, Februrary 1984.

....as printf) within an agent. A similar package, Mediating Connectors [3] also traps Windows library procedures, but permits the composition of multiple agents. The rules used by this package are compared with those of Bypass above in section 4.2. Remote procedure calls are a standard technique [5,17,15]. Our facility is similar to other implementations, but is driven by the need for drop in software which works without modifying the target application. To this end, our specification syntax relies on annotating existing interfaces instead of creating new protocols from scratch. Likewise, the RPC ....

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, Februrary 1984.

....over the network. The two additional requirements are that the server instance must register its existence and location with the RPC system, and the client must identify the target server to the RPC system. These operations are typically performed once, when the client or server is first started [1]) Not only does RPC make creating distributed services easier from the programmer s perspective, but its widespread usage have led developers to produce ever more efficient implementations of RPC, to the point where the cost of RPC is very small compared to the latency inherent in even the ....

....from using existing network bandwidth more efficiently. One thought would be to complement GIOP with a set of inter ORB protocols with different invocation semantics, such as a true one way invocation based on UDP, a connectionless, request response mechanism (which the original RPC system used [1]) and UDP based multicast. A potential workaround to enable CORBA applications to use UDP would be to develop a UDP sender receiver combination. CORBA could be used for all reliable operations (including sender and receiver set up) while the UDP sender receiver combination could be used for ....

Andrew D. Birrell and Bruce Jay Nelson. Implementing Remote Procedure Calls. ACM Trans. On Computer Systems 2(1), February 1984, pp. 39-59.

....of RPC MBCF and chapter5 shows performance evaluation of RPC MBCF. chapter6 is a conclusion of this thesis. 3 Chapter 2 Remote Procedure Calls 2.1 Remote Procedure Call Remote Procedure Call (RPC) is a software library for remote execution. RPC was first introduced by Birrel and Nelson in 1984[2]. With RPC,user can call a procedure on remote host in the same way of calling a local procedure. A RPC library hides message packing unpacking and transferring from user. In general RPC client hides these processing form user. 1. User calls a local procedure with server name. This procedure call ....

Andrew Birrell and Bruce Nelson. Implementing remote procedure calls. ACM Trans. Computer Systems, 2(1):39--59, February 1984.

.... stdio.h #include netdb.h #define SERV TCP PORT 4555 client for socket based server int main (int argc, char argv[ 5 int sockfd,i; char c[100] struct sockaddr in server; struct hostent host info; char server name; char buffer[100] server name = argv[1] buffer[0] argv[2][0] for (i=3;i argc;i )memcpy( buffer[1 16 (i 3) argv[i] 16) if( sockfd=socket(AF INET, SOCK STREAM, 0) 0) printf( can t open client socket n ) exit(1) host info = gethostbyname(server name) if(host info = NULL) fprintf(stderr, s: unknown host: s n , argv[0] ....

....behaviour of the server. An alternative method for writing distributed applications which handles most of these issues automatically and in which the interaction between the program parts is much less visible is called Remote Procedure Calls (RPC) and was rst proposed by Birrel and Nelson. [2] The intention of all RPC proposals is to make programming for a distributed solution to a problem as close as possible to that of a non distributed solution. In particular, communication between the server and the client should have the appearance of normal procedure calls. This goal was only ....

[Article contains additional citation context not shown here]

Birrel, A. D. and Nelson, B. J. (1984) Implementing Remote Procedure Calls, ACM Transactions on Computer Systems, vol.2, no.1, pp. 39-59, Feb. 1984

....close them. If both are expensive, as is often the case, the transport can cache idle connections for a limited amount of time. It is perfectly possible to implement a class of connection that communicates with datagrams according to a protocol that makes idle connections essentially free see [3]. That is, in spite of its name, the type Connection need not be connection oriented in the standard sense of the word. A connection c contains a reader c. rd and a writer c.wr. Connections come in pairs; if c and d are paired, whatever is written to c. wr can be read from d.rd, and vice versa. ....

Andrew D. Birrell and Bruce Jay Nelson. Implementing Remote Procedure Call. A CSJ Transactions o r Co mpurer Systems 2(1), February 1984, pp. 39-59.

No context found.

A. D. Birrel and B. J. Neslon. Implementing Remote Procedure Calls. ACM Trans. on Computer Systems, 2:39--59, 1984.

No context found.

A. Birrel and B. Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1), February 1984.

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing Remote Procedure Calls. ACM Transactions on Computer Systems, 2(1):39--59, February 1984.

No context found.

Birrel, A. D. and B. J. Nelson, Implementing remote procedure calls, ACM Transactions on Computer Systems 2 (1984), pp. 39--59.

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2:39-59, 1984.

No context found.

Birrel, A. and B. Nelson. "Implementing remote procedure calls", ACM Transactions on Computer Systems 2(1), Feb 1984, pp.39-59.

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, Februrary 1984.

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems (TOCS), 2(1):39--59, 1984.

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, Februrary 1984.

No context found.

A. Birrel and B. Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1), February 1984.

No context found.

A.D. Birrel and B.J. Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, February 1984.

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, February 1984.

No context found.

A. Birrel and B. Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1), February 1984.

No context found.

A. Birrel and B. Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1), February 1984.

No context found.

Andrew D. Birrel and Bruce J. Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, February 1984.

No context found.

Birrel, A. D., and Nelson, B. J., "Implementing Remote Procedure Calls", XEROX CSL-83-7, October 1983.

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39-59, February 1984.

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2(1):39--59, February 1984. 39.

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing Remote Procedure Calls. In Proceedings of the 9th ACM Symposium on Operating Systems Principles (SOSP'83). October 1983.

No context found.

Andrew Birrell and Bruce Nelson. Implementing remote procedure calls. ACM Trans. Computer Systems, 2(1):39--59, February 1984.

No context found.

A.D.Birrel, B.J.Nivat `Implementing remote procedure calls' ACM Transactions on Computer Systems 2, 1 (February 1984) 39-59.

No context found.

Andrew Birrell and Bruce Nelson, `Implementing remote procedure calls', ACM Trans. Computer Systems, 2(1), 39--59 (February 1984).

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing remote procedure calls. ACM Transactions on Computer Systems, 2#1#:39#59, February 1984.

No context found.

Andrew D. Birrell & Bruce Jay Nelson. "Implementing Remote Procedure Calls". ACM Transactions on Computer Systems, Vol. 2 No. 1, pp. 39-59. 1984

No context found.

Andrew D. Birrell and Bruce Jay Nelson. Implementing Remote Procedure Calls. In ACM Transactions on Computer Systems, Vol. 2, No. 1, February 1984.

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