F. Bodin, P. Beckman, D. Gannon, and others. "Distributed pC++: Basic Ideas for an Object Parallel Language", Scientific Programming, Volume 2/Number 3 Fall 1993  Home/Search   Document Details and Download   Summary   Related Articles   Check

.... object array is quite distinct from the array objects found in HPF, POOMA[13] P [14] Global Arrays[12] and elsewhere. In our construct, each element of the array is a relatively coarse grained 1 C object, with full support for remote method invocation. Our work is quite similar to pC [8], but adds migration and reductions. Unlike Concurrent Aggregates[9] Linda[7] or Orca[10] there is no duplication or replication message sends address exactly one array element across the entire machine. This work is complemented by fast collective communication libraries such as [11] but ....

F. Bodin, P. Beckman, D. Gannon, and others. "Distributed pC++: Basic Ideas for an Object Parallel Language", Scientific Programming, Volume 2/Number 3 Fall 1993

....into a single global variable, they provide support for accumulators and monotonic variables , as well as distributed data tables (similar in concept to Linda) and branch office chares , which have a server on each processor to handle certain requests locally. 2.1. 2 pC pC (Bodin et al. [7]) is closer to what we are trying to do, in the sense that its parallel extensions are based on adding additional data structure constructors to the language rather than using a process based approach. pC defines collections, which are regular data structures, in particular arrays, whose ....

Francois Bodin, Peter Beckman, Dennis Gannon, Srinivas Narayana, and Shelby Yang. "Distributed pC++: Basic Ideas for an Object Parallel Language ". Obtained electronically from Indiana University Department of Computer Science, 1992.

....is an extension to Fortran 77 that supports loop and functional parallelisms in message passing programs. Extensions to Fortran 90 to support task (functional) parallelism are discussed in [33] Languages with data distribution features are not limited to dialects of Fortran. Distributed pC [34] adds parallel and data distribution constructs to C . Other parallel versions of C include Composition C [35] C [36] COOL [37] and CHARM [38] The pC (parallel C) programming language [39] is a parallel extension to ANSI C. Concurrent Aggregates (CA) 40, 41] is a parallel language ....

F. Bodin, P. Beckmann, D. Gannon, S. Narayana, and S. X. Yang, "Distributed pC++: Basic ideas for an object parallel language," tech. rep., Indiana University, 1994.

....distributed memory, MIMD 1. Introduction The object oriented paradigm has proven to be a powerful tool for managing complexity in the development of software for sequential computers, and its power is being exploited in the more complex domain of software for parallel machines as well [5] 6][8][22] 32] 40] Software for parallel machines must provide efficient run time support if it is to achieve acceptable performance. This is particularly true for distributed memory MIMD machines where communication and synchronization costs can be large. Given that high performance is the raison ....

....dependencies as they develop, rather than at compile time. This facilitates larger grain computation by eliminating many small grain control actors which are inefficient in a distributed memory environment. Second, Mentat supports both task and data parallelism, not just data parallelism as in [8][15] 18] 28] 33] Third, Mentat operates over a spectrum of architectures, from loosely coupled heterogeneous networks of workstations, to tightly coupled multicomputers. Finally, Mentat has a scalable, distributed, control mechanism. This includes both the mechanism used to construct and modify ....

[Article contains additional citation context not shown here]

F. Bodin, et. al., "Distributed pC++: Basic Ideas for an Object Parallel Language," Proceedings Object-Oriented Numerics Conference, pp. 1-24,Sunriver, Oregon, April 25-27, 1993.

.... What has changed that makes the realization of a complete high performance metacomputer possible The change is that achieving high performance via parallelism, previously available only for tightly coupled parallel processors, is now possible for loosely coupled distributed systems [1] 3] 4][6][8] 14] 27] 33] 37] 45] Whether or not a metasystem is explicitly constructed by design, the nation (and perhaps the world) will eventually build a system that shares at least some of the attributes of Legion. The reason is simple: individual and organizational users will be required to deal with ....

F. Bodin, et. al., "Distributed pC++: Basic Ideas for an Object Parallel Language," Proceedings Object-Oriented Numerics Conference, pp. 1-24,Sunriver, Oregon, April 25-27, 1993.

.... Fortran 77D [Bozkus:93a] Fox:91e] Hiranandani:92c] Vienna Fortran [Chapman:92b] C [Hatcher:91a;91b] Crystal [Chen:88b] APL; Fortran for SIMD parallel computers ffl Loosely Synchronous: Extensions of the above, especially HPF [Chapman:94b] Choudhary:92d] HPF:94a] and parallel C [Bodin:91a], Chandy:93a] Grimshaw:93b] Lemke:92a] Fortran or C plus message passing [Fox:91m] McBryan:94a] ffl Asynchronous: Linda [Factor:90a;90b] Gelertner:89a] CC [Chandy:93a] Time Warp [Wieland:89a] PCN [Chandy:90a] WebWork [Fox:95a] ffl Compound Metaproblems: AVS [Mills:92a;92b] ....

Bodin, F., Beckman, P., Gannon, D., Narayana, S., and Shelby, Y. "Distributed pC++: Basic ideas for an object parallel language," in Proceedings of Supercomputing '91, pages 273--282. (IEEE) Computer Society and (ACM) (SIGARCH), November 1991.

.... Fortran 77D [Bozkus:93a] Fox:91e] Hiranandani:92c] Vienna Fortran [Chapman:92b] C [Hatcher:91a;91b] Crystal [Chen:88b] APL; Fortran for SIMD parallel computers ffl Loosely Synchronous: Extensions of the above, especially HPF [Chapman:94b] Choudhary:92d] HPF:94a] and parallel HPC [Bodin:91a], Chandy:93a] Grimshaw:93b] Lemke:92a] Fortran or C plus message passing [Fox:91m] McBryan:94a] ffl Asynchronous: Linda [Factor:90a;90b] Gelertner:89a] CC [Chandy:93a] Time Warp [Wieland:89a] PCN [Chandy:90a] WebWork [Fox:95a] including Java based distributed simulation ffl ....

Bodin, F., Beckman, P., Gannon, D., Narayana, S., and Shelby, Y. "Distributed pC++: Basic ideas for an object parallel language," in Proceedings of Supercomputing '91, pages 273--282. (IEEE) Computer Society and (ACM) (SIGARCH), November 1991.

....object oriented, parallel processing 1. Introduction The object oriented paradigm has proven to be a powerful tool for managing complexity in the development of software for sequential computers, and its power is being exploited in the more complex domain of parallel software as well [3] 4][6][14] 22] 23] In parallel systems, where performance is a key objective, the issue of run time support must be addressed, particularly for distributed memory MIMD machines. For years the object oriented approach has had a reputation for high run time overhead and poor performance. Recent ....

F. Bodin, et. al., "Distributed pC++: Basic Ideas for an Object Parallel Language," Proceedings Object-Oriented Numerics Conference, pp. 1-24,Sunriver, Oregon, April 25-27, 1993.

....which provides control parallelism at the method level. Data parallel applications have been implemented using this language[9] However, many of the techniques used to develop these applications are automatically provided in current data parallel languages such as 2 Dataparallel C[11] and pC [2]. A great deal of handcoding was required to achieve the same result with the MPL. We believe this lack of support is a deficiency of the language. There are a number of data parallel languages in existence today. Our extensions build on the previous research done to develop these languages, ....

....of iteration code to loop over contained elements, and other mind numbing details. Details of the computation model (macro data flow) the Mentat programming language and the Mentat run time system can be found elsewhere [7, 8, 17] 2. 3 Related Work Dataparallel C [10, 11, 18, 19] pC [2, 15] C [14] Fortran D [5] Fortran 90 [10] and High Performance Fortran (HPF) 16] are the languages from which we have borrowed ideas and which are related to our work. We have also developed some new ideas. C and pC are based on C . Dataparallel C is based on C, but uses some ideas from ....

[Article contains additional citation context not shown here]

F. Bodin et al., "Distributed pC++: Basic Ideas for an Object Parallel Language," Proceedings Object-Oriented Numerics Conference, April 25-27, 1993, Sunriver, Oregon, pp. 1-24.

....designed to utilize the data parallel capabilities of the CM 2 and CM 5. In a similar manner, Cedar Fortran (Hoeflinger, 1991) was designed to utilize the clusters of the Cedar architecture. More generic languages have also been developed such as HPF (High Performance Fortran Forum, 1993) and pC (Bodin et al. 1993) which do not target a specific machine, but instead target a class of parallel machines. These types of languages make it easier to utilize the capabilities of the architecture, but they do not provide any additional understanding of the algorithms or the application under development. A problem ....

Bodin, F. et al., 1993. Distributed pC++: Basic ideas for an object parallel language. In OON-SKI'93 Proc. of the 1st Ann. Object-Oriented Numerics Conf. pp. 1--24.

.... realization of a complete high performance metacomputer possible What has changed is that solutions to the problem of providing high performance via parallelism, previously available only for tightly coupled parallel processors, are now available for loosely coupled distributed systems [1] 3] 4][8][9] 12] 21] 26] 29] 36] This change in the technological landscape makes possible the realization of practical high performance metasystems. Whether or not a metasystem is explicitly constructed by design, the nation (and probably at some time the world) will eventually end up with a system that ....

F. Bodin, et. al., "Distributed pC++: Basic Ideas for an Object Parallel Language," Proceedings Object-Oriented Numerics Conference, pp. 1-24,Sunriver, Oregon, April 25-27, 1993.

....and polymorphism can add a new value to the building process of large parallel applications. All of the four discussed languages make a use of objects. Orca [2] is an object based language which does not support inheritance but allows object s reusability and hierarchical strutters. CC [3] , pC [9] and Charm [11] are truly object oriented languages, which are based on the most popular objectoriented language C [12] The first section presents a short description of the languages. We are comparing different approaches and features of these languages in the second section. ....

....However, remote references do not provide any synchronization between participating nodes. In CC one should use single assignment variables and atomic functions to synchronize and in pC MethodOfElement methods (they provide an implicit barrier synchronization and the end of each function [3]) There are cases when the synchronization at the end of a function is not enough. In that moment, a programmer should use the Barrier call. pC does not provide any synchronization tools more refined than barriers. The Charm implementation is quite different from the two languages above. ....

F. Bodin, P. Beckman, D. Gannon, S. Narayana, S. Yang, "Distributed pC++: Basic Ideas for an Object Parallel Language", Technical Report.

....permit the explicit representation of parallelism, where they are necessary for the construction of compilers that generate target specific source code containing calls to run time systems that support concurrency. Examples include the Fortran D and Fortran 90D compilers [1] and the pC compiler [5]. Source tosource translation systems are also used to construct compilers that tackle the problem of discovering parallelism in sequential source code. These compilers apply restructuring transformations to generate a version of the program that uses directives or explicit parallel constructs ....

....added to the Sage hierarchy. The unparser module, which is table driven, must also be extended to recognize this new node. While we have done this several times (we have added some of the PCF, Fortran S, Fortran M and HPF extensions to Fortran and extended C to define our pC language [5] as well as the proposed CC syntax) it is not an easy task, because it requires a complete understanding of the internal parser structures. A future version of Sage will work with a different parser generator that will, we hope, simplify this task. 7 Conclusions The Sage object hierarchy ....

F. Bodin, D. Gannon, P. Beckman, S. Narayana, S. Yang , "Distributed pC++: Basic Ideas for an Object Parallel Language," In Proceedings of the First Annual Object-Oriented Numerics Conference (OON-SKI), Sunriver, Oregon, Apr. 1993, pp. 1-24.

No context found.

F. Bodin, P. Beckman, D. Gannon, S. Narayana, and S. Yang, `Distributed pC++: Basic ideas for an object parallel languages', Report, Indiana University, (Jan 1993).

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