On Hardware-Agnostic Programming for Parallel Architectures

<jats:p>Writing programs for parallel systems is difficult and time consuming. Typically, a separate program has to be written for each platform, and the different processors and memories have to be coordinated manually. These challenges make it important to build programming languages that can abstract away from these hardware details. This dissertation contributes to these high-level languages, in particular for distributed-memory systems. To ensure good quality scientific codes, we need to optimise the overall structure. We also show that the high-level feature rank-polymorphism can assist in this.</jats:p>