Collaboration Strategies for Fog Computing under Heterogeneous Network-bound Scenarios

摘要: The success of IoT applications increases the number online devices and motivates adoption a fog computing paradigm to support large widely distributed infrastructures. However, heterogeneity nodes their connections requires introduction load balancing strategies guarantee efficient operations. This aspect is particularly critical when some are characterized by high communication delays. Some proposals such as Sequential Forwarding algorithm have been presented in literature provide systems. algorithms not studied for wide range working parameters an heterogeneous infrastructure; furthermore, these designed take advantage from highly network delays that common contribution this study twofold: first, we evaluate performance sequential forwarding several delay conditions; second, propose test delay-aware version takes into account presence variable node connectivity infrastructure. results our experiments, carried out using realistic topology, demonstrate delay-blind approach may determine poor represents major response time. Furthermore, show variant benefit case, with reduction time up 6%.