Programming heterogeneous architectures using hierarchical tasks

Abstract

SummaryTask‐based systems have become popular due to their ability to utilize the computational power of complex heterogeneous systems. A typical programming model used is the Sequential Task Flow (STF) model, which unfortunately only supports static task graphs. This can result in submission overhead and a static task graph that is not well‐suited for execution on heterogeneous systems. A common approach is to find a balance between the granularity needed for accelerator devices and the granularity required by CPU cores to achieve optimal performance. To address these issues, we have extended the STF model in the StarPU runtime system by introducing the concept of hierarchical tasks. This allows for a more dynamic task graph and, when combined with an automatic data manager, it is possible to adjust granularity at runtime to best match the targeted computing resource. That data manager makes it possible to switch between various data layout without programmer input and allows us to enforce the correctness of the DAG as hierarchical tasks alter it during runtime. Additionally, submission overhead is reduced by using large‐grain hierarchical tasks, as the submission process can now be done in parallel. We have shown that the hierarchical task model is correct and have conducted an early evaluation on shared memory heterogeneous systems using the Chameleon dense linear algebra library.

Enhancing Sparse Direct Solver Scalability Through Runtime System Automatic Data Partition

“…However, this strategy, when pushed to its limit, generates a large number of tasks that may be dicult to process eciently by the scheduler. In a future work, we plan to explore the possibility given by StarPU's hierarchical tasks [6] to take dynamically the decision to rene, or not, the granularity. Additionally, it would help to delay the submission of ne-grained tasks to reduce the overhead of the scheduler.…”

Performance Portability of Generated Cardiac Simulation Kernels Through Automatic Dimensioning and Load Balancing on Heterogeneous Nodes

“…Regarding task granularity management, some works aggregate multiple fine-grained task into one coarse-grained task [21]. Other studies take the opposite approach of starting from coarse-grained and splitting them into finer-grained task when appropriate [8], [25]. Granularity management, starting from a DSL and generating code using STARPU runtime have been studied in the particular context of linear algebra [7].…”