Optimizing Array Accesses in High Productivity Languages

Abstract

Abstract. One of the outcomes of DARPA's HPCS program has been the creation of three new high productivity languages: Chapel, Fortress, and X10. While these languages have introduced improvements in language expressiveness and programmer productivity, several technical challenges still remain in delivering high performance with these languages. In the absence of optimization, the high-level language constructs that improve productivity can result in order-of-magnitude runtime performance degradations. This paper addresses the problem of efficient code generation for high level array accesses in the X10 language. Two aspects of high level array accesses in X10 are important for productivity but also pose significant performance challenges: the high level accesses are performed through Point objects rather than integer indices, and variables containing references to arrays are rank-independent. Our solution to the first challenge is to extend the X10 compiler with automatic inlining and scalar replacement of Point objects. Our partial solution to the second challenge is to use X10's dependent type system to enable the programmer to annotate array variable declarations with additional information for the rank and region of the variable, and to allow the compiler to generate efficient code in cases where the dependent type information is available. Although this paper focuses on high level array accesses in X10, our approach is applicable to similar constructs in other languages. Our experimental results for single-thread performance demonstrate that these compiler optimizations can enable high-level X10 array accesses with implicit ranks and Points to improve performance by up to a factor of 5.4× over unoptimized X10 code, and to also achieve performance comparable (from 48% to 100%) to that of lower-level Java programs. These results underscore the importance of the optimization techniques presented in this paper for achieving high performance with high productivity.

Array optimizations for parallel implementations of high productivity languages

“…One consequence of the point-wise for loop in the X10 version is that (by default) it leads to an allocation of a new point object in every iteration for the index and for all subscript expressions, thereby significantly degrading performance. We previously addressed this problem (for sequential execution only) with a point inlining optimization [13]. However, after applying this transformation, we still experience up to 2 orders of magnitude in performance degradation when comparing Java Grande benchmarks with X10's general high-level arrays against the same benchmarks with lower-level Java arrays.…”

“…We also describe an array transformation strategy (Section 3.4), that uses the results from our rank analysis algorithm to convert general X10 arrays into a lower-level, more efficient Java arrays. These two techniques, combined with object inlining of points [5,12,13] result in performance improvements of up to two orders of magnitude. In Section 4, we validate our techniques on a set of parallel Java Grande benchmarks [11].…”

Subregion Analysis and Bounds Check Elimination for High Level Arrays

“…The first two challenges have been addressed in the past [17,18]. In this paper, we address the optimizing the overhead of array bounds checks by developing a novel regionbased interprocedural array bounds analysis to automatically identify redundant checks.…”