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

Mobile Edge Computing (MEC) has emerged as a promising supportingarchitecture providing a variety of resources to the network edge, thus actingas an enabler for edge intelligence services empowering massive mobile andInternet of Things (IoT) devices with AI capability. With the assistance ofedge servers, user equipments (UEs) are able to run deep neural network (DNN)based AI applications, which are generally resource-hungry andcompute-intensive, such that an individual UE can hardly afford by itself inreal time. However the resources in each individual edge server are typicallylimited. Therefore, any resource optimization involving edge servers is bynature a resource-constrained optimization problem and needs to be tackled insuch realistic context. Motivated by this observation, we investigate theoptimization problem of DNN partitioning (an emerging DNN offloading scheme) ina realistic multi-user resource-constrained condition that rarely considered inprevious works. Despite the extremely large solution space, we reveal severalproperties of this specific optimization problem of joint multi-UE DNNpartitioning and computational resource allocation. We propose an algorithmcalled Iterative Alternating Optimization (IAO) that can achieve the optimalsolution in polynomial time. In addition, we present rigorous theoreticanalysis of our algorithm in terms of time complexity and performance underrealistic estimation error. Moreover, we build a prototype that implements ourframework and conduct extensive experiments using realistic DNN models, whoseresults demonstrate its effectiveness and efficiency.

Joint Multi-User DNN Partitioning and Computational Resource Allocation for Collaborative Edge Intelligence