A.S. Grimshaw. An introduction to parallel object-oriented programming with Mentat. Technical Report 91 07, University of Virginia, 1991.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....is sometimes because it doesn t support the operations (operator overloading, for example) and sometimes because it does the work for you (automatic garbage collection, for example) In recent years numerous variations on the theme of C for parallel computing have appeared. See, for example [8, 17, 23, 28, 31, 55]. 21 2.4.3 C# C# [20] which is an important part of the new Microsoft .NET platform, is a modern, object oriented and type safe programming language based on C and C . Header files, Interface Definition Language, and complicated interfaces are not needed in C#. C# is supposed to o#er an ....

A.S. Grimshaw. An introduction to parallel object-oriented programming with Mentat. Technical Report 91 07, University of Virginia, 1991.

....parallel computations without reducing their flexibility [20] The master slave model forms the basis of the AERO system [3] Unlike n:i. gen , however, the model is com pletely asynchronous with the master having to explicitly accept results from parallel slave computations. The Mentat system [15] is based on an object oriented programming language derived from C . The user encapsulates data and computation as objects and the compiler performs all the necessary data flow analysis to permit parallel invocations whenever possible. The run time system ensures correct object semantics even ....

A. S. Grimshaw. An Introduction to Parallel Object-Oriented Programming with Mentat. Technical Report No. TR-91-07, Department of Computer Science, University of Virginia, April 1991.

.... Java is a descendant of C . C and C are used increasingly in scienti c programming; they are already used almost universally by implementers of parallel libraries and compilers. In recent years numerous variations on the theme of C for parallel computing have appeared. See, for example [7, 28, 11, 2, 12, 21]. 3 Java omits various features of C and C that are considered dicult notably, pointers. Poor compiler analysis has often been blamed on these features. The inference is that Java, like Fortran, may be a suitable source language for highly optimizing compilers (although direct evidence ....

....message passing and data parallelism. As observed in the previous section, communication through true shared memory is already implicit in the Java thread model. Communication through remote objects is undoubtedly a natural and important paradigm in Java, especially for access to remote services [27, 22, 21], but we will not discuss it further here. An orthogonal issue is whether the high level libraries used to implement these paradigms (presuming class library implementations) should be written in Java on top of the standard API, or whether they should be implemented as direct interfaces to native ....

A.S. Grimshaw. An introduction to parallel object-oriented programming with Mentat. Technical Report 91 07, University of Virginia, 1991. 22

....some reasons to think that Java may be a good language for scientific and parallel programming. Java is a descendant of C . C and C are used increasingly in scientific programming. In recent years numerous variations on the theme of C for parallel computing have appeared. See, for example [26, 6, 9, 2, 10, 18]. Java omits various features of C and C that are considered difficult notably, pointers. Poor compiler analysis has often been blamed on these features. The inference is that Java, like Fortran, may be a suitable source language for highly optimizing compilers (although direct evidence ....

....message passing and data parallelism. As observed in the previous section, communication through true shared memory is already implicit in the Java thread model. Communication through remote objects is undoubtedly a natural and important paradigm in Java, especially for access to remote services [24, 19, 18], but it is not specifically tied to scientific parallel programming and we will not discuss it further here. 3 Message passing case studies Message passing remains one of the most effective and widely used communication paradigms in parallel computing. In this section we compare two approaches ....

A.S. Grimshaw. An introduction to parallel object-oriented programming with Mentat. Technical Report 91 07, University of Virginia, 1991.

....large number of projects studying C for parallel hardware. Our work can be distinguished from these projects by our design goals: ffl Support strongly the data parallel paradigm. ffl Tolerate no changes to C . ffl Allow automatic generation of implementations for parallel hardware. Mentat [7] and CC [4] are both dialects of C designed to exploit parallel hardware. Both languages are most effective as SPMD programming environments. Neither language strongly supports the data parallel model as they require the programmer to explicitly perform synchronization and communication ....

A. Grimshaw. An introduction to parallel object-oriented programming with Mentat. Technical Report 91--07, University of Virginia, 1991.

....Templates Library such as the Amelia Vector Template Library (AVTL) 6] While C is the most common language for this work, it is not the only way object parallelism may be expressed. There are many other parallel programming models that exploit object oriented concepts. These include Mentat[7], CC [8] CORBA[9] Charm [10] and UC [11] In these systems, the emphasis is on task level parallelism on networked 2 and heterogeneous systems. As will be explained later, the run time architectures for these systems have different requirements than for the data parallel extensions ....

A. S. Grimshaw. An introduction to parallel object-oriented programming with Mentat, Technical Report 91 07, University of Virginia, 1991.

....the response has not returned by the time that the program requires the value, the program blocks while waiting for the value to be returned. The variable, declared as a promise or future, is able to be used as a parameter in other functions that return future variables. Legion, based on Mentat [12] structures computations in terms of a macro dataflow model. The system uses a mechanism whereby the future value of a function may be used as a parameter in another function. Mentat is aimed at instruction level parallelism rather than at the coarse grained parallelism of the DISCWorld project. ....

Andrew S. Grimshaw. An Introduction to Parallel ObjectOriented Programming with Mentat. Technical Report CS-9107, University of Virginia, April 1991.

....concept. The following categorization is not mutually exclusive a language may suitably fall into more than one category. To reduce the amount of survey, we give the design characteristics of the following languages in Table 2. 1, but omit any description Natasha [78] Amber [64] Mentat [115, 116, 183], PROCOL [224] CLIX [141] Dragoon [19] Sloop [167] Hybrid [184] PRESTO (and other systems which provides threads and synchronization libraries [94, 33, 27, 90] SOS [206, 169] Rosette [217] POOL2 [8, 9, 10] We have however described some of their more interesting characteristics earlier. ....

Andrew S. Grimshaw. An Introduction to the Parallel Object-Oriented Programming with Mentat. Technical Report TR-91-07, Department of Computer Science, University of Virginia, Charlottesville, Virginia 22903, April 1991.

....because the total storage for an object is usually composed of several sub units. For example, a matrix object may contain storage for each row. In this case, each row would become a memory node. All the rows together make up the composite memory node for the entire object. Mentat Mentat [14] is a parallel object oriented programming system that uses the distributed memory approach for objects. It is based on the object oriented sequential programming language C . The goal of the Mentat system is to provide the programmer with efficient, easy to use parallelism in an object oriented ....

....they are all ad hoc. VMPP has the advantage that it is a general solution and also provides a better solution in most cases. 6.1. Single Language Solutions to the Portability Problem There are currently several parallel languages designed to provide portability across parallel architectures [8, 14]. The approach taken by these languages is to provide a single parallel programming language along with compilers for a variety of parallel architectures. The language is usually designed to be relatively architecture independent. The single language approach suffers from two problems. First, ....

Andrew S. Grimshaw, "An Introduction to Parallel Object-Oriented Programming with Mentat", Computer Science Report No. TR-91-07, Department of Computer Science, University of Virginia, Charlottesville, VA, April 4, 1991.

....of data dependencies; the overall model is one of task parallelism. Jade has been implemented as an extension of the C programming language on shared and distributed memory machines and on heterogeneous workstation clusters. 2.6. 10 Mentat Mentat, developed at the University of Virginia [45], 46] is an environment for coarsegrain parallel application development. The Mentat language is derived from C and includes support for explicitly identified parallel class types. One of the more interesting aspects of Mentat is the run time tracking and enforcement of data dependencies. The ....

A. S. Grimshaw, "An introduction to parallel object-oriented programming with Mentat," Department of Computer Science, University of Virginia, Charlottesville, VA, Tech. Rep. TR-91-07, 1991.

....30 E.1 A Better Matrix : 30 1 Introduction 1. 1 Existing Parallel Machines and Languages There are many different parallel languages available today for a variety of parallel architectures and models; examples include Mentat[7], PC [8] Modula 3 [9] and Split C[10] Using these languages one can soon be overwhelmed by their complexity and variance: The languages are different, often vastly so. Few are designed to conform to a model such as Valiant s BulkSynchronous Parallel[1] model. Parallel languages are often ....

....with support for Abstract Data Types. We can achieve this using object oriented programming. As object oriented parallel languages already exist, I must first justify the need for a new one. The most important reason is that those which do exist are complicated and not BSP oriented. DoPVM, Mentat[7] and PC [8] are three commonly cited languages: These are essentially different (as is my language) DoPVM is a fundamentally different approach to parallelism and is not suited to BSP. Mentat and PC are both SPMD with automatic compiler included synchronisation. Mentat and PC are similar to ....

[Article contains additional citation context not shown here]

A. Grimshaw, University of Virginia, An Introduction to Parallel Object-oriented Programming with Mentat, April 1991.

....between the two. These mixed systems offer finer grained distribution than a user s process, but they do not usually work with the kernel itself. Instead, they sit on top of the kernel and use a run time model other than a UNIX process, such as an object. Systems of this type include Mentat[4], Linda[5] and SPEDE. Remote procedure calls [6] can also be considered in this category, although they are only part of an integrated environment (for example, they do not support remote process execution) Linda [5] is a unique distributed environment because it uses a much different paradigm ....

....in when a result is found. This is clearly different from SPEDE, which is based on objects sending messages to other objects. While the tuple space model is powerful because of its simplicity and abstractness, performance may be limited due to the complexities of managing the tuple space. Mentat [4], on the other hand, uses an object oriented paradigm based on the C programming language. Language extensions to C allow the system to handle remote communication and synchronization. For instance, after a remote invocation, the caller does not block until the return value from the invocation ....

A. Grimshaw. An Introduction to Parallel Object-Oriented Programming with Mentat. UVA Computer Science Report No. TR-91-07. April 4, 1991.

.... BETA [20] Ori 5 ent 84 [17] and POOL T [2] None of these languages have been implemented on a parallel machine 5 [25, page 5] Since 1989, five attempts at implementing the Actor model on various parallel architectures have taken place: # Hal [16] operates on Sequent s Symmetry; # Mentat [10] operates on BBN Butterfly and Sun workstations; # Ellie [3] operates on the Transputer network and Sun workstations. # ACT [18] operates on Sequent s Symmetry; # ASK [9] operates under MS DOS and on a Transputer network. These experiments, however, gave rise only to prototypes intended ....

A.S. Grimshaw, An introduction to parallel object-oriented programming with Mentat, Computer Science Report No. TR-91-07, University of Virginia, Avril 1991.

....runtime systems and communication libraries based on RPC. For example, Nexus and Legion are both highly portable runtime systems that support a variety of user level communication protocols and have been used as a compiler target for parallel languages such as CC [8] Fortran M [13] and Mentat [16]. Both of these systems are aimed for wide area, large scale parallel computing. These languages and systems implement RPC on top of user level messaging layers such as MPI, PVM, or even TCP sockets. While the implementation is tuned for performance, maintaining portability, flexibility, and ....

A. Grimshaw. An Introduction to Parallel Object-Oriented Programming with Mentat, Technical Report 91-07, University of Virginia, July 1991.

....parallel computations without reducing their flexibility [38] The master slave model forms the basis of the AERO system [3] Unlike nigen , however, the model is completely asynchronous with the master having to explicitly accept results from parallel slave computations. The Mentat system [26] is based on an object oriented programming language derived from C . The user encapsulates data and computation as objects and the compiler performs all the necessary data flow analysis to permit parallel invocations whenever possible. The run time system ensures correct object semantics even ....

A. S. Grimshaw. An Introduction to Parallel Object-Oriented Programming with Mentat. Technical Report No. TR-91-07, Department of Computer Science, University of Virginia, April 1991.

....failure detection and recovery. It is clear that effective use of distributed systems for parallel computing requires automated support in handling these complexities. A number of projects are under way to harness the power of distributed systems and support network based parallel computing [2,3,8,9,14,12]. The Paralex programming environment is an instance of such a system. Paralex takes graphical descriptions of parallel computations and automatically produces fault tolerant distributed programs to carry them out. The system is a complete programming environment in that it includes a graphics ....

A. S. Grimshaw. An Introduction to Parallel Object-Oriented Programming with Mentat. Technical Report No. TR-91-07, Department of Computer Science, University of Virginia, April 1991.

....reasons to think that Java may be a good language for scientific and parallel programming. ffl Java is a descendant of C . C and C are used increasingly in scientific programming. In recent years numerous variations on the theme of C for parallel computing have appeared. See, for example [25, 6, 9, 2, 10, 17]. ffl Java omits various features of C and C that are considered difficult notably, pointers. Poor compiler analysis has often been blamed on these features. The inference is that Java, like Fortran, may be a suitable source language for highly optimizing compilers (although direct ....

....message passing and data parallelism. As observed in the previous section, communication through true shared memory is already implicit in the Java thread model. Communication through remote objects is undoubtedly a natural and important paradigm in Java, especially for access to remote services [23, 18, 17], but it is not specifically tied to scientific parallel programming and we will not discuss it further here. 3 Message passing case studies Message passing remains one of the most effective and widely used communication paradigms in parallel computing. In this section we compare two approaches ....

A.S. Grimshaw. An introduction to parallel object-oriented programming with Mentat. Technical Report 91 07, University of Virginia, 1991.

....Collaboratory systems Tango[3] 1.1 DOGMA 1.1.1 Relation of Previous Work to DOGMA DOGMA seeks to do more than simply provide a Java version of a message passing standard. Rather, the philosophy of DOGMA is to allow object oriented parallel programming somewhat similar to that offered by Mentat[12], but in a system independent way. Like Mentat, DOGMA attempts to strike a middle ground between ease of programming and system performance. DOGMA differs mainly from current parallel Java systems in that it is an object group oriented system designed mainly for running on clusters of workstations ....

A. S. Grimshaw. An introduction to parallel object-oriented programming with mentat. Technical Report 91-07, University of Virginia, 1991.

....and the parallel composition operator in PCN [CT91] The use of par blocks differs from most other concurrent object oriented languages in that with a par block, multiple threads of control exist within a single object. Other languages tend to associate thread creation with object creation [BLL88, Gri91, SS87, Agh86] Consequently, only one thread of control is ever associated with an object. The spawn statement in CC can be used to achieve the same effect. The advantage of par blocks and parfor statements over the tying thread creation to object creation approach is twofold. First, these ....

Andrew S. Grimshaw. An introduction to parallel object-oriented programming with Mentat. Computer Science Report TR-91-07, University of Virginia, 1991.

....communication and integrated performance monitoring. It promotes the exploitation of heterogeneity, as opposed to attempting to hide it. The design of HBD has also been influenced by a number of other systems; e.g. Agora [5] Marionette [13] HIGHLAND [15] Concert [20] PHARROS [8] and Mentat [9]. 4.1 Motivational Analogy To motivate the design of the HBD programming environment, let us make an analogy to human systems. Many of the most difficult and important open research problems cross disciplines (e.g. medicine, biology, and computer science) In order to solve such problems, it is ....

A. Grimshaw. An Introduction to Parallel Object-Oriented Programming with Mentat. University of Virginia Technical Report. TR-91-07. 1991.

....to program and, hence, inaccessible to mainstream, production system programmers. The Mentat approach exploits the object oriented paradigm to provide high level abstractions that mask the complex aspects of parallel programming, communication, synchronization, and scheduling, from the programmer [2, 3]. Instead of managing these details, the programmer concentrates on the 3 application. The programmer uses application domain knowledge to specify those object classes that are of sufficient computational complexity to warrant parallel execution. The parallelization of these complex tasks is ....

A. S. Grimshaw, "An Introduction to Parallel Object-Oriented Programming with Mentat," TR-91-07, Department of Computer Science, University of Virginia, April, 1991.

No context found.

A. S. Grimshaw. An introduction to parallel object-oriented programming with Mentat. Technical Report TR-91-07, University of Virginia, Charlottesville, Va., 1991. 6

No context found.

A.S. Grimshaw. An introduction to parallel object-oriented programming with Mentat. Technical Report 91-07, Computer Science Department, University of Virginia, April 1991.

No context found.

Andrew S. Grimshaw. An introduction to parallel object-oriented programming with Mentat. Technical Report TR-91-07, University of Virginia Computer Science, 1991.

No context found.

Andrew S. Grimshaw. An Introduction to Parallel Object Oriented Programming with MENTAT. Technical report, University Of Virginia, Charlottsville, VA 22903, April 1991.

First 50 documents

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