Spar: a set of extensions to Java for scientific computation

Abstract

SUMMARYWe present a set of language extensions that improve the expressiveness and performance of Java for scientific computation. The language extensions allow the manipulation of multi-dimensional arrays to be expressed more naturally, and to be implemented more efficiently. Furthermore, data-parallel programming is supported, allowing efficient parallelization of a large class of operations on arrays. We also provide language extensions to construct specialized array representations, such as symmetric, block, and sparse matrices. These extensions are: tuples, parameterized types, array subscript overloading, and the inline modifier. These extensions are not only useful in the construction of special array representations, but are also useful in their own right. Finally, we add complex numbers as a primitive type to the language. We evaluate our language extensions using performance results. We also compare relevant code fragments of our extended language with standard Java implementations and language extensions proposed by others.

Using MPI with C# and the Common Language Infrastructure

“…The latter are essential for efficient representation of "small objects" such as complex numbers, coordinate vectors, rotation matrices, etc. Similar improvements for Java have been proposed by Java researchers [1,13,15,20]. Standard C# has these now.…”

Adding tuples to Java

“…The set of extensions was chosen to substantially improve the support of Java for scientific computations. For further information see [13]. The Timber compiler is available for downloading at [12].…”

“…For example: int a[*,*] = new int [9,9] causes 12 to be assigned to array element a [1,1]. For more information on this aspect of our language extensions see [13].…”

Java for Scientific Computation: Prospects and Problems

“…Two such solutions are proposed in the Spar/Java project [14] and the Titanium project [16]. For example, a two-dimensional array in Spar/Java is declared and used as follows:…”

“…Spar [14] provides a foreach loop that specifies that the iterations of the loop can be executed in arbitrary order, but once an iteration is started, it must be completed before the next iteration can be started. Arrays can be annotated with HPF-like distribution pragmas to indicate on which processor the data must be stored.…”