Wholey, S., and Steele Jr., G. L. Connection Machine Lisp: A dialect of CommonLisp for data parallel programming. In Proceedings Second International Conference on Supercomputing (May 1987).  Home/Search   Document Not in Database   Summary   Related Articles   Check

....parallel programming once students have been taught C. In this paper, we assume that the reader is knowledgeable of ANS C. The concepts designed into aCe C have incorporated features found in parallel languages such APL[3] DAP Fortran [4] Parallel Pascal[5] Parallel Forth[6] C [7] CM lisp [8], FGPC [9] and MPL[10] aCe has been implemented on top of native C compilers such as gcc and Maspar MPL and on top of message passing libraries such as Cray SHMEM, MPI [11] and PVM [12] In this paper, we will focus on particular features of aCe C (from now on will refer to it as aCe) with ....

....the programmer to both envision the most logical architecture for the application and then implement the algorithm using that architecture. Many instructors of parallel computing courses like to have their students program in a variety of parallel programming models such as data parallel [4] 7[8][9] 10] message passing (MPI) 11] Pthreads [20] and concurrent sequential processing [21] The programmer assumes that statements are executed in lock step and the compiler optimizations and the runtime system take the liberty to synchronize only when it is necessary. This usually happens ....

Steele, G., Wholey, S., "Connection Machine Lisp: A Dialect of Common Lisp for Data Parallel Programming," August, 1987.

....in parallel with the others. Clearly the traditional execution model underlying flat data parallelism cannot provide such parallelism within parallelism . Thus, to directly support nested data parallelism we need to consider generalisations of that execution model. Several language systems [23, 17] which allow nested data parallel specifications choose to solve this problem by arbitrarily serialising all but one dimension of parallelism. This effectively re establishes the property that only a single data parallel operation may be active within the system at any time. Thus such ....

S. Wholey and G. Steele Jr. Connection machine lisp: A dialect of Common Lisp for data parallel programming. In Proceedings of the Second International Conference on Supercomputing, 1987. y Copies of these technical reports can be found at URL: ftp://ftp.cs.adelaide.edu.au/pub/ParFP/techreports

....languages support are restricted to simple collection data types like arrays or similar structures. Examples includes imperative data parallel languages like C [RS87] Dataparallel C [HQ91] and High Performance Fortran [For93] and functional data parallel languages like Connection Machine Lisp [WS94] and NESL [Ble92] Therefore, some important parallel algorithms based on other data structures like trees would become rather awkward. For example, trees are indirectly represented using several vectors in NESL, and the parallel algorithms on trees become difficult to understand. Secondly, the ....

S. Wholey and Jr. G. L. Steele. Connection machine Lisp: A dialect of common Lisp for data parallel programming. In ACM Symposium on Parallel Algorithms and Architectures, June 1994.

....beginning of each segment, the lengths specifies the length of each segment, and the head pointers points to the beginning of each segment. 14. To solve many problems in computational geometry [4] 15. To efficiently implementlanguages that support nested parallel structures [5] such as CM LISP [23], on flat parallel machines. Comparison with previous work There has been significant research on parallel and vector algorithms for unsegmented scans. A parallel implementation of scans on a perfect shuffle network was suggested by Stone [19] for polynomial evaluation. The problem with this ....

Skef Wholey and Guy L. Steele Jr. Connection Machine Lisp: A Dialect of Common Lisp for Data Parallel Programming. In Proceedings Second International Conference on Supercomputing, May 1987.

....descriptor manipulation instructions, which create and manipulate segment descriptors; and I O instructions, which read and write vectors and segment descriptors. The permute instructions can be used to implement vector based indexing in C , Fortran 90, and APL, for the fi operation in CM Lisp [63], and for the move operation in Paralation Lisp [51] The segmented instructions permit the implementation of the nested parallelism allowed in CM Lisp, SETL, and Paralation Lisp. The rest of the VCODE instructions are for flow control, stack management and system interface. Of special note are ....

.... UC [5] and Fortran 90 [2] Section 2 explained some of the expressibility and efficiency limitations imposed by flat languages; these problems are also discussed elsewhere [10, 11, 29, 32, 54] Two existingparallel languages permit the user to describe nested data parallel operations: CM Lisp [63] and Paralation Lisp [51] However, the implementations of these languages only exploit the bottom level of parallelism; for the sparse matrix example, this results in a parallel sum for each row and a serial loop over the rows. Both of these languages are data parallel extensions to Common Lisp ....

Skef Wholey and Guy L. Steele Jr. Connection Machine Lisp: A dialect of Common Lisp for data parallel programming. In Proceedings Second International Conference on Supercomputing, May 1987.

....unstructured sparse matrices, graphs or trees [21] Languages with control parallel constructs are often better suited for such problems, but unfortunately these constructs do not port well to vector machines, SIMD machines or MIMD machines with vector processors. Nested data parallel languages [45, 7, 22] combine aspects of both strict data parallel languages and controlparallel languages. Nested data parallel languages allow recursive data structures, and also the application of parallel functions to multiple sets of data in parallel. For example, a sparse array can be represented as a sequence ....

....Section 2 explained some of the expressibility and efficiency limitations imposed by this type of language. These problems are also discussed elsewhere [7, 21, 39, 22, 8] There are two existing languages that permit the user to describe nested data parallel operations: Connection Machine Lisp [45] and Paralation Lisp [36] However, the implementations of these languages are only able to exploit the bottom level of parallelism; for the sparse matrix example, this results in a parallel sum for each row, and a serial loop over the rows. Both these languages are data parallel extensions to ....

S. Wholey and G. L. Steele Jr. Connection Machine Lisp: A dialect of Common Lisp for data parallel programming. In Proceedings Second International Conference on Supercomputing, May 1987.

....for these structures in a conventional programming language. Examples include C 3 [RS87] Dataparallel C [HQ91] 3 Lisp [Las86] and High Performance Fortran [For93] There are several recent proposals that integrate data parallelism in a functional calculus, including Connection Machine Lisp [WS87], Paralation Lisp [Sab88] TUPLE [Yua92] DPML [HF93] and Nesl [Ble93] However, parallelism in these proposals is still limited to simple collection data types. From this current situation, one may think that data parallelism would be inherently restricted to applications manipulating arrays. The ....

S. Wholey and Guy L. Steele Jr. Connection machine lisp: A dialect of common lisp for data parallel programming. In Proc. International Conference on Supercomputing, May 1987.

....be over committed, and thus no scheduling is required. These two schemes for avoiding the complexity of the general proposal for a nested data parallel model of execution have been used in a number of existing systems (e.g. for nested FORALL statements in CM Fortran [14] Connection Machine Lisp [19], Paralation Lisp [12] The techniques essentially revert the requirements of the nested paradigm to those that can be easily satisfied by the traditional paradigm. There is, however, a cost in applying either of these methods, namely a potential loss in parallelism. While a rigorous analysis of ....

S. Wholey and Guy L. Steele Jr. Connection machine lisp: A dialect of Common Lisp for data parallel programming. In Proceedings of the Second International Conference on Supercomputing, 1987.

No context found.

Wholey, S., and Steele Jr., G. L. Connection Machine Lisp: A dialect of CommonLisp for data parallel programming. In Proceedings Second International Conference on Supercomputing (May 1987).

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