DawnCC

Mendonca, Gleison Souza Diniz; Guimarães, Breno C. F.; Alves, Péricles; Pereira, Marcio Machado; Araújo, Guido; Pereira, Fernando Magno Quintão

doi:10.1145/3084540

Cited by 25 publications

(4 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compile-time setting of accesses: Many research efforts target automatic compile-time parallelization, e.g., Cetus [24], DawnCC [25], [26], AutoPar [27], Pluto [28] and TaskMiner [29]. These primarily target data parallelism, and only TaskMiner is specific to tasks.…”

Section: Related Workmentioning

confidence: 99%

Automatic aggregation of subtask accesses for nested OpenMP-style tasks

Shaaban

Aguilar

Beltrán

et al. 2022

2022 IEEE 34th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD)

View full text Add to dashboard Cite

Task-based programming is a high performance and productive model to express parallelism. Tasks encapsulate work to be executed across multiple cores or offloaded to GPUs, FPGAs, other accelerators or other nodes. In order to maintain parallelism and afford maximum freedom to the scheduler, the task dependency graph should be created in parallel and well in advance of task execution. A key limitation with OpenMP and OmpSs-2 tasking is that a task cannot be created until all its accesses and its descendents' accesses are known. Current approaches to work around this limitation either stop task creation and execution using a taskwait or they substitute "fake" accesses known as sentinels. This paper proposes the auto clause, which indicates that the task may create subtasks that access unspecified memory regions or it may allocate and return memory at addresses that are of course not yet known. Unlike approaches using taskwaits, there is no interruption to the concurrent creation and execution of tasks, maintaining parallelism and the scheduler's ability to optimize load balance and data locality. Unlike existing approaches using sentinels, all tasks can be given a precise specification of their own data accesses, so that a single mechanism is used to control task ordering, program data transfers on distributed memory and optimize data locality, e.g. on NUMA systems. The auto clause also provides an incremental path to develop programs with nested tasks, by removing the need for every parent task to have a complete specification of the accesses of its descendent tasks. This is redundant information that can be time consuming and error-prone to describe. We present a straightforward runtime implementation that achieves a 1.4 times speedup for n-body with OmpSs-2@Cluster task offloading to 32 nodes and <4% slowdown for three benchmarks with task offloading to 8 nodes. All code is open source.

show abstract

Section: Related Workmentioning

confidence: 99%

Automatic aggregation of subtask accesses for nested OpenMP-style tasks

Shaaban

Aguilar

Beltrán

et al. 2022

2022 IEEE 34th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD)

View full text Add to dashboard Cite

show abstract

“…CCAMP [24] is an OpenACC and OpenMP interoperability framework; interoperability in the context of CCAMP consists of source-to-source transformation of one model to the other for the sake of portability rather than the joint execution and collaboration of both models. Finally, other compiler centric work such as DawnCC [25], automates the insertion of OpenACC data copy directives and generates optimized host-device data transfers; however, it does not involve multi-GPU kernel scheduling in its scope.…”

Section: Related Workmentioning

confidence: 99%

OmpSs-2 and OpenACC Interoperation

Korakitis¹,

Gonzalo

Guidotti

et al. 2022

2022 Workshop on Accelerator Programming Using Directives (WACCPD)

View full text Add to dashboard Cite

We propose an interoperation mechanism to enable novel composability across pragma-based programming models. We study and propose a clear separation of duties and implement our approach by augmenting the OmpSs-2 programming model, compiler and runtime system to support OmpSs-2 + OpenACC programming. To validate our proposal we port ZPIC, a kinetic plasma simulator, to leverage our hybrid OmpSs-2 + OpenACC implementation. We compare our approach against OpenACC versions of ZPIC in terms of on a multi-GPU HPC system. We show that our approach manages to provide automatic asynchronous and multi-GPU execution, removing significant burden from the application's developer, while also being able to outperform manually programmed versions, thanks to a better utilization of the hardware.

show abstract

“…Mendonça et al 26 introduce DawnCC, a source-to-source compiler built on top of the LLVM framework to automatically insert OpenMP and OpenAcc offloading pragmas for loops that can be parallelized. As its main contribution, DawnCC automatically extracts bounds of memory regions and uses these bounds to for the data movement primitives of OpenMP and OpenAcc.…”

Section: Related Workmentioning

confidence: 99%

Heterogeneous computing with OpenMP and Hydra

Diener

Kalé

Bodony

2020

Concurrency and Computation

View full text Add to dashboard Cite

Summary High‐performance computing relies on accelerators (such as GPGPUs) to achieve fast execution of scientific applications. Traditionally, these accelerators have been programmed with specialized languages, such as CUDA or OpenCL. In recent years, OpenMP emerged as a promising alternative for supporting accelerators, providing advantages such as maintaining a single code base for the host and different accelerator types and providing a simple way to extend support for accelerators to existing application codes. Efficiently using this support requires solving several challenges, related to performance, work partitioning, and concurrent execution on multiple device types. In this article, we discuss our experiences with using OpenMP for accelerators and present performance guidelines. We also introduce a library, Hydra, that addresses several of the challenges of using OpenMP for such devices. We apply Hydra to a scientific application, PlasCom2, that has not previously been able to use accelerators. Experiments on three architectures show that Hydra results in performance gains of up to 10× compared with CPU‐only execution. Concurrent execution on the host and GPU resulted in additional gains of up to 20% compared to running on the GPU only.

show abstract

DawnCC

Cited by 25 publications

References 37 publications

Automatic aggregation of subtask accesses for nested OpenMP-style tasks

Automatic aggregation of subtask accesses for nested OpenMP-style tasks

OmpSs-2 and OpenACC Interoperation

Heterogeneous computing with OpenMP and Hydra

Contact Info

Product

Resources

About