Energy-aware dynamic task mapping for NoC-based MPSoCs

Mandelli, Marcelo; Ost, Luciano; Carara, Everton Alceu; Guindani, Guilherme; Gouvea, Thiago; Medeiros, Guilherme Cardoso; Moraes, Fernando

doi:10.1109/iscas.2011.5937903

Cited by 26 publications

(12 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The MPSoC used to evaluate the mapping heuristics is configured as follows: 2D-mesh topology, XY routing algorithm, 16 Table 2 presents the total execution time to execute 10 application iterations, for each evaluated scenario, considering up to three tasks mapped to the same PE. The average execution time overhead compared to the single-task LEC-DN, is 14%, 13%, 16% and 19% for the NN, BN, LEC-DN multi-task and Premap-DN heuristics respectively.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Multi-task dynamic mapping onto NoC-based MPSoCs

Mandelli

Amory

Ost

et al. 2011

Proceedings of the 24th Symposium on Integrated Circuits and Systems Design

Self Cite

View full text Add to dashboard Cite

Task mapping defines the best placement of a given task in the MPSoC, according to some criteria, as energy or Manhattan distance minimization. The ITRS roadmap forecast in a near future MPSoCs with hundreds of processing elements (PEs). Therefore, dynamic mapping heuristics are required. An important gap is observed in the mapping literature: the lack of proposals targeting multi-task dynamic mapping. In this context, the present work proposes an energy-aware dynamic task mapping heuristic, allowing multiple tasks allocation per PE. Experimental results are executed in an actual MPSoC running distributed applications. Comparing a single-task to the multi-task mapping, the energy spent in the NoC is reduced in average by 51% (best case: 72%), with an average execution time overhead of 18%. Besides the communication energy reduction, the multi-task mapping enables a greater number of applications executing simultaneously, or smaller MPSoCs, which reduces the system cost.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Differently from NN and BN heuristics, which map the target task as close as possible to its source task, the LEC-DN [16] considers the proximity of the target task to all communicating tasks that are already mapped. LEC-DN employs two search methods to select the PE that receives the target task.…”

Section: Lec-dnmentioning

confidence: 99%

Multi-task dynamic mapping onto NoC-based MPSoCs

Mandelli

Amory

Ost

et al. 2011

Proceedings of the 24th Symposium on Integrated Circuits and Systems Design

Self Cite

View full text Add to dashboard Cite

show abstract

“…In turn, Mandelli et al [87] propose a power-aware task mapping heuristic, which is validated using a NoC-based MPSoC described at a cycle-accurate level. The mapping heuristic is performed in a given processor of the system that executes a preemptive operating system.…”

Section: Allocation Techniques For Energy-efficiencymentioning

confidence: 99%

Dynamic Resource Allocation in Embedded, High-Performance and Cloud Computing

Indrusiak

Dziurzański

Singh

2016

Dynamic Resource Allocation in Embedded, High-Performance and Cloud Computing

View full text Add to dashboard Cite

The "River Publishers Series in Information Science and Technology" covers research which ushers the 21st Century into an Internet and multimedia era. Multimedia means the theory and application of filtering, coding, estimating, analyzing, detecting and recognizing, synthesizing, classifying, recording, and reproducing signals by digital and/or analog devices or techniques, while the scope of "signal" includes audio, video, speech, image, musical, multimedia, data/content, geophysical, sonar/radar, bio/medical, sensation, etc. Networking suggests transportation of such multimedia contents among nodes in communication and/or computer networks, to facilitate the ultimate Internet.Theory, technologies, protocols and standards, applications/services, practice and implementation of wired/wireless networking are all within the scope of this series. Based on network and communication science, we further extend the scope for 21st Century life through the knowledge in robotics, machine learning, embedded systems, cognitive science, pattern recognition, quantum/biological/molecular computation and information processing, biology, ecology, social science and economics, user behaviors and interface, and applications to health and society advance.Books published in the series include research monographs, edited volumes, handbooks and textbooks. The books provide professionals, researchers, educators, and advanced students in the field with an invaluable insight into the latest research and developments.

show abstract

“…At the device level, low power distribution of tasks among the cores of an MPSoC device is not a new topic [4] Heuristic and optimal solutions have been proposed considering the energy consumption of the cores or interconnect network between them.…”

Section: Related Workmentioning

confidence: 99%

Unified energy aware mapping of multimedia applications on media cloud infrastructure

Moghadas

Hashemi

2014

2014 22nd Iranian Conference on Electrical Engineering (ICEE)

View full text Add to dashboard Cite

Media clouds provide an infrastructure for distributed processing and storage of multimedia applications. Each device in the media cloud is usually a mobile system consisting of several heterogeneous Processing Elements (PE). Energy efficiency has always been a challenge in this environment. In this paper, we propose a unified energy efficient algorithm for mapping multimedia applications on a media cloud infrastructure. The algorithm considers the distribution of tasks at two levels: among devices in the cloud and PEs of each device. The proposed algorithm heuristically finds the most energy efficient device for each task at the first stage. In the second stage, the least energy consuming PE among all available PEs is chosen for mapping. The efficiency of the proposed scheme has been evaluated by mapping H.264/AVC decoder on a media cloud. Simulation results indicate that by offloading application tasks on more power efficient PEs, the overall energy consumption for this application would be reduced by up to 12 times in a typical scenario in comparison to local processing. Meanwhile, the performance of the heuristic has been compared to the optimal solution.

show abstract

Energy-aware dynamic task mapping for NoC-based MPSoCs

Cited by 26 publications

References 9 publications

Multi-task dynamic mapping onto NoC-based MPSoCs

Multi-task dynamic mapping onto NoC-based MPSoCs

Dynamic Resource Allocation in Embedded, High-Performance and Cloud Computing

Unified energy aware mapping of multimedia applications on media cloud infrastructure

Contact Info

Product

Resources

About