An Integrated Hardware-Software Approach to Task Graph Management

Engelhardt, Nina; Dallou, Tamer; Elhossini, Ahmed; Juurlink, Ben

doi:10.1109/hpcc.2014.66

Cited by 6 publications

(10 citation statements)

References 13 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [11], the integration process is highlighted with the multicore RTS, and the evaluation of Nexus++ with real applications.…”

Section: Nexus++ Hardware Task Managermentioning

confidence: 99%

“…Although Nexus++ has one central task manager, it has demonstrated significant improvement over the software RTS using trace-based simulations [11]. Nevertheless, Nexus++ could not improve the scalability of the H264dec benchmark over the software version, since it doesn't support the barrier pragma taskwait on.…”

Section: A Nexus++ Processing Pipelinementioning

confidence: 99%

“…The RTS checks all inputs and outputs for tasks submitted for execution and then when all dependencies are resolved, tasks are marked as ready and submitted to one of the free cores for execution. The dependency resolution is shown to be a bottleneck in the RTS of task based programming models such as StarSs [14], and OmpSs [11]. This bottleneck affects the scalability of applications on multi-core systems as no significant speed up can be achieved by increasing the number of processing cores to run parallel tasks.…”

Section: Introductionmentioning

confidence: 99%

“…Building custom hardware accelerators is one of the solutions that was investigated in literature to reduce the overhead of dependency resolution and increase the overall scalability of parallel applications. The hardware accelerator introduced in [7] named Nexus++ and analyzed in detail in [11], presented a significant enhancement in terms of scalability for various applications. Nexus++ is used to speed up the dependency resolution in OmpSs's RTS.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Nexus#: A Distributed Hardware Task Manager for Task-Based Programming Models

Dallou

Engelhardt

Elhossini

et al. 2015

2015 IEEE International Parallel and Distributed Processing Symposium

Self Cite

View full text Add to dashboard Cite

Abstract-In the era of multicore systems, it is expected that the number of cores that can be integrated on a single chip will be 3-digit. The key to utilize such a huge computational power is to extract the very fine parallelism in the user program. This is nontrivial for the average programmer, and becomes very hard as the number of potential parallel instances increases. Task-based programming models such as OmpSs are promising, since they handle the detection of dependencies and synchronization for the programmer. However, state-of-the-art research shows that task management is not cheap, and introduces a significant overhead that limits the scalability of OmpSs. Nexus# is a hardware accelerator for the OmpSs runtime system, which dynamically monitors dependencies between tasks. It is fully synthesizable in VHDL, and has a distributed task graph model to achieve the best scalability. Supporting tasks with arbitrary number of parameters and any dependency pattern, Nexus# achieves better performance than Nanos, the official OmpSs runtime system, and scales well for the H264dec benchmark with very fine grained tasks, among other benchmarks from the Starbench suite.

show abstract

“…In [11], the integration process is highlighted with the multicore RTS, and the evaluation of Nexus++ with real applications.…”

Section: Nexus++ Hardware Task Managermentioning

confidence: 99%

Section: A Nexus++ Processing Pipelinementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Nexus#: A Distributed Hardware Task Manager for Task-Based Programming Models

Dallou

Engelhardt

Elhossini

et al. 2015

2015 IEEE International Parallel and Distributed Processing Symposium

Self Cite

View full text Add to dashboard Cite

show abstract

“…To overcome this deficiency and enable a finer task parallelism, We propose to improve the runtime by offloading some of the most time consuming runtime functions [3] (dependence analysis and task scheduling) to hardware. Our work on hardware task dependence graph management has showed great scalability and performance improvement over its software-only alternatives [2], [4], [5].…”

Section: Introductionmentioning

confidence: 99%

Picos, A Hardware Task-Dependence Manager for Task-Based Dataflow Programming Models

Tan

Bosch

Vidal

et al. 2017

2017 International Conference on High Performance Computing &Amp; Simulation (HPCS)

View full text Add to dashboard Cite

Abstract-Task-based programming Task-based programming models such as OpenMP, Intel TBB and OmpSs are widely used to extract high level of parallelism of applications executed on multi-core and manycore platforms. These programming models allow applications to be expressed as a set of tasks with dependences to drive their execution at runtime. While managing these dependences for task with coarse granularity proves to be highly beneficial, it introduces noticeable overheads when targeting fine-grained tasks, diminishing the potential speedups or even introducing performance losses. To overcome this drawback, we propose a hardware/software co-design Picos that manages inter-task dependences efficiently. In this paper we describe the main ideas of our proposal and a prototype implementation. This prototype is integrated with a parallel taskbased programming model and evaluated with real executions in Linux embedded system with two ARM Cortex-A9 and a FPGA. When compared with a software runtime, our solution results in more than 1.8x speedup and 40% of energy savings with only 2 threads.

show abstract