English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Fog/Edge computing is a novel computing paradigm supportingresource-constrained Internet of Things (IoT) devices by the placement of theirtasks on the edge and/or cloud servers. Recently, several Deep ReinforcementLearning (DRL)-based placement techniques have been proposed in fog/edgecomputing environments, which are only suitable for centralized setups. Thetraining of well-performed DRL agents requires manifold training data whileobtaining training data is costly. Hence, these centralized DRL-basedtechniques lack generalizability and quick adaptability, thus failing toefficiently tackle application placement problems. Moreover, many IoTapplications are modeled as Directed Acyclic Graphs (DAGs) with diversetopologies. Satisfying dependencies of DAG-based IoT applications incuradditional constraints and increase the complexity of placement problems. Toovercome these challenges, we propose an actor-critic-based distributedapplication placement technique, working based on the IMPortance weightedActor-Learner Architectures (IMPALA). IMPALA is known for efficient distributedexperience trajectory generation that significantly reduces the explorationcosts of agents. Besides, it uses an adaptive off-policy correction method forfaster convergence to optimal solutions. Our technique uses recurrent layers tocapture temporal behaviors of input data and a replay buffer to improve thesample efficiency. The performance results, obtained from simulation andtestbed experiments, demonstrate that our technique significantly improves theexecution cost of IoT applications up to 30\% compared to its counterparts.

A Distributed Deep Reinforcement Learning Technique for Application Placement in Edge and Fog Computing Environments