Real-time co-scheduling of multiple dataflow graphs on multi-processor systems

Kang, Shin-haeng; Kang, Duseok; Yang, Hoeseok; Ha, Soonhoi

doi:10.1145/2897937.2898077

Cited by 7 publications

(12 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The b-level of tasks in H a are shown in Table 3. Thus, execution order of tasks on CORE1 and CORE2 are denoted as {v 1 7 Based on the given priority and v/f assignment for each task, a scheduling with PMM should be generated to reduce total energy consumption. The time interval can be directly modeled as the following:…”

Section: Lpesmentioning

confidence: 99%

“…Thus, they can be adjusted (swapped) as long as the precedence constraints are guaranteed. The corresponding schedule after FT can be seen in Figure 5b, execution order of tasks on CORE1 and CORE2 are {v 1 7…”

Section: Ftmentioning

confidence: 99%

“…The multiprocessor architecture of an MPSoC incorporates multiprocessors and other functional units in a single case on a single die. Meanwhile, there is an ongoing trend that diverse emerging safety-critical real-time applications, such as automotive, computer vision, data collection and control applications are running simultaneously on the MPSoCs [1]. For these safety-related applications, it is imperative that deadlines should be strongly guaranteed.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Energy-Efficient Scheduling of Periodic Applications on Safety-Critical Time-Triggered Multiprocessor Systems

et al. 2018

View full text Add to dashboard Cite

Abstract:Energy optimization for periodic applications running on safety/time-critical time-triggered multiprocessor systems has been studied recently. An interesting feature of the applications on the systems is that some tasks are strictly periodic while others are non-strictly periodic, i.e., the start time interval between any two successive instances of the same task is not fixed as long as task deadlines can be met. Energy-efficient scheduling of such applications on the systems has, however, been rarely investigated. In this paper, we focus on the problem of static scheduling multiple periodic applications consisting of both strictly and non-strictly periodic tasks on safety/time-critical time-triggered multiprocessor systems for energy minimization. The challenge of the problem is that both strictly and non-strictly periodic tasks must be intelligently addressed in scheduling to optimize energy consumption. We introduce a new practical task model to characterize the unique feature of specific tasks, and formulate the energy-efficient scheduling problem based on the model. Then, an improved Mixed Integer Linear Programming (MILP) method is proposed to obtain the optimal scheduling solution by considering strict and non-strict periodicity of the specific tasks. To decrease the high complexity of MILP, we also develop a heuristic algorithm to efficiently find a high-quality solution in reasonable time. Extensive evaluation results demonstrate the proposed MILP and heuristic methods can on average achieve about 14.21% and 13.76% energy-savings respectively compared with existing work.

show abstract

Section: Lpesmentioning

confidence: 99%

Section: Ftmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Energy-Efficient Scheduling of Periodic Applications on Safety-Critical Time-Triggered Multiprocessor Systems

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Since the SDF graph describes only the data dependency between tasks, it can be easily parallelized by mapping task instances to PEs. While the parallel scheduling of a single SDF graph onto heterogeneous PEs has been extensively researched ( [3], [4], [5], [6]), there exist only a few studies that consider the scheduling of multiple SDF graphs on the shared PEs ( [7], [8], [9]). One approach to schedule multiple SDF graphs is to merge all SDF graphs into a single SDF graph after expanding each of them to the hyperperiod and schedule the merged graph at once [7].…”

Section: Introductionmentioning

confidence: 99%

Parallel Scheduling of Multiple SDF Graphs Onto Heterogeneous Processors

et al. 2021

Self Cite

View full text Add to dashboard Cite

Parallel scheduling of multiple real-time applications onto heterogeneous processors is needed in the emerging embedded systems such as self-driving cars, smart cameras, and smartphones. Assuming that an embedded application is specified as a synchronous dataflow (SDF) graph or its extension, we propose a novel parallel scheduling methodology based on an evolutionary algorithm where the mapping of tasks onto processors is evolved to optimize a given objective function in an iterative fashion. In each iteration, we use an existing worst-case response time (WCRT) analysis tool to check if all applications satisfy their real-time requirements by translating each SDF graph into a directed acyclic graph (DAG) that is assumed in the WCRT analysis tool. Since the WCRT analysis must be performed in each iteration of evolution, we propose a clustering technique to reduce drastically the analysis time that depends on the number of nodes and their dependency. We formally prove that the proposed clustering technique does not change the estimated WCRT of each application. The effectiveness of the proposed scheduling methodology with the clustering technique is verified with extensive experiments using real-life benchmarks, randomly generated graphs, and the comparison with the existing technique.

show abstract

“…With the rapid development of grid computing and cloud computing applications, in a heterogeneous distributed computing environment on how to improve the multiple task scheduling performance aspects put forward the new requirements, also caused the wide attention of scholars both at home and abroad [8][9]. For multiprocessor task scheduling, how to improve the throughput of multi-task scheduling system is a research hotspot [10]. At present, the researchers have proposed many about methods of improving the performance of DAG task scheduling, but most of these articles focus on how to improve the efficiency of without time constraints of scheduling, in order to complete all the tasks as fast as possible.…”

Section: Introductionmentioning

confidence: 99%

A novel HEFTS-L algorithm for scheduling large number of DAG tasks

Liu¹,

Xu²,

Cheng³

et al. 2017

Proceedings of the 3rd Workshop on Advanced Research and Technology in Industry (WARTIA 2017)

View full text Add to dashboard Cite

Abstract:In recent years, along with the development of technologies for distributed computing such as big data and clouds workflow systems, efficiency of workflow scheduling has become very impotent. Hence scheduling of multiple DAGs sharing on heterogeneous distributed resources has attracted intensive attention recently. This paper issues on scheduling of multiple DAGs with Deadline constraints surrounding scheduling of multiple DAGs sharing on heterogeneous distributed resources on traditional DAG-based task. It brings forward two parameter association methods to balance DAG Deadline constraint priority including relative strictness and laxity, which are used to balance priority of multiple DAGs. An improved HEFT algorithm based on priority was put forward to schedule multiple DAGs with deadline constraint. Through experiments, the phenomenon of "overload" can be detected due to the high degree of emergency of DAG scheduling. Once a task was detected, it can be selective discarded with corresponding measures, so as to maximize DAG before deadline.

show abstract

Real-time co-scheduling of multiple dataflow graphs on multi-processor systems

Cited by 7 publications

References 6 publications

Energy-Efficient Scheduling of Periodic Applications on Safety-Critical Time-Triggered Multiprocessor Systems

Energy-Efficient Scheduling of Periodic Applications on Safety-Critical Time-Triggered Multiprocessor Systems

Parallel Scheduling of Multiple SDF Graphs Onto Heterogeneous Processors

A novel HEFTS-L algorithm for scheduling large number of DAG tasks

Contact Info

Product

Resources

About