Visible to the public A Diffusional Schedule for Traffic Reducing on Network-on-Chip

TitleA Diffusional Schedule for Traffic Reducing on Network-on-Chip
Publication TypeConference Paper
Year of Publication2018
AuthorsWang, Jihe, Zhang, Meng, Qiu, Meikang
Conference Name2018 5th IEEE International Conference on Cyber Security and Cloud Computing (CSCloud)/2018 4th IEEE International Conference on Edge Computing and Scalable Cloud (EdgeCom)
Keywordsapplication-layer optimization, Coherence, diffusional pattern, diffusional schedule, energy consumption, energy efficient task schedule, energy saving, energy., inefficient task schedule strategy, latency, latency reduction, Load modeling, long-distance data synchronization, Metrics, network on chip, network on chip security, network-on-chip, NoC activities, parallel processing, parallel task execution, power aware computing, Protocols, pubcrawl, remote data access, resilience, Resiliency, Scalability, Schedules, scheduling, synchronisation, Synchronization, Task Analysis, task mapping algorithm, task schedule, task traffic reduction, traffic distribution
Abstractpubcrawl, Network on Chip Security, Scalability, resiliency, resilience, metrics, Tasks on NoC (Network-on-Chip) are less efficient because of long-distance data synchronization. An inefficient task schedule strategy can lead to a large number of remote data accessing that ruins the speedup of parallel execution of multiple tasks. Thus, we propose an energy efficient task schedule to reduce task traffic with a diffusional pattern. The task mapping algorithm can optimize traffic distribution by limit tasks into a small area to reduce NoC activities. Comparing to application-layer optimization, our task mapping can obtain 20% energy saving and 15% latency reduction on average.
Citation Keywang_diffusional_2018