PLDI

This paper aims to foster cross-fertilisation between programming language and compiler research performed on different array programming language infrastructures. We study how to enable better comparability of concepts and techniques by looking into ...

We present the data parallel programming library Petalisp --- an extension of Common Lisp for data parallel programming. The core of Petalisp is deliberately simple. It features only a single data structure --- the strided array --- and four operations ...

In recent years, MATLAB's just-in-time (JIT) interpreter has improved the execution time of for-loops to the extent that loops can outperform equivalent array operations in some scenarios. This has caused systematic translation of loops to array ...

Array algorithms where operations are applied to disjoint parts of an array lend themselves well to parallelism, since parallel threads can operate on the parts of the array without synchronisation. However, implementing such algorithms requires ...

Rank polymorphism serves as a type of control flow used in array-oriented languages, where functions are automatically lifted to operate on high-dimensional arguments. The iteration space is derived directly from the shape of the data, presenting a ...

Software correctness in general is a hard problem, and especially so for high performance computing (HPC). One problem being that array layout and traversal may depend on array size and hardware properties (cache size, core count, etc), making ...

Previous work has shown how the well-studied and SIMD-friendly Structure of Arrays (SOA) data layout strategy can speed up applications in high-performance computing compared to a traditional Array of Structures (AOS) data layout. However, a standard ...

As we move towards exascale computing, computer architecture is bound to see dramatic changes. Multiple nodes, with or without shared memory, multicore and accelerators (GPUs, FPGAs) will be the norm. For many domains, such as finite difference ...