Workload‐aware loop schedulers were introduced to deliver better performance than
classical loop scheduling strategies. However, they presented limitations such as inflexible …

Supercomputers today employ a large number of cores on each node, often with additional
symmetric multi-threading on each core [1, 13]. At the same time, load imbalance is …

Performance irregularities on massively parallel processors lead to load imbalances and a
significant loss of performance. Multi-core nodes suggest a promising way to redistribute …

Many different sources of overheads impact the efficiency of a scheduling strategy applied to
a parallel loop within a scientific application. In prior work, we handled these overheads …

The first part of the presentation will motivate why novel loop scheduling schemes are
needed for scientific applications. As applications and supercomputer architectures are …

Supercomputers today employ a large number of cores on each node, often with additional
symmetric multi-threading on each core. The Charm++ parallel programming system …