Programming Heterogeneous Architectures Using Hierarchical Tasks

Task-based systems have gained popularity as they promise to exploit the computational power of complex heterogeneous systems. A common programming model is the so-called Sequential Task Flow (STF) model, which, unfortunately, has the intrinsic limitation of supporting static task graphs only. This leads to potential submission overhead and to a static task graph not necessarily adapted for execution on heterogeneous systems. A standard approach is to find a trade-off between the granularity needed by accelerator devices and the one required by CPU cores to achieve performance. To address these problems, we extend the STF model of StarPU  [5] to enable tasks subgraphs at runtime. We refer to these tasks as hierarchical tasks. This approach allows for a more dynamic task graph. Combined with an automatic data manager, it allows to dynamically adapt the granularity to meet the optimal size of the targeted computing resource. We show that the model is correct and we provide an early evaluation on shared memory heterogeneous systems, using the Chameleon [1] dense linear algebra library.