Title :
Distributed load balancing for resilient information-centric SeDAX networks
Author :
Hoefling, Michael ; Mill, Cynthia G. ; Menth, Michael
Author_Institution :
Commun. Networks, Univ. of Tuebingen, Tubingen, Germany
Abstract :
SeDAX is a publish/subscribe information-centric networking architecture where publishers send messages to the appropriate message broker over a Delaunay-triangulated overlay network. Resilient data forwarding and data redundancy enable a high level of reliability. Overlay nodes and topics are addressed via geo-coordinates. A topic is stored on primary and secondary nodes, those nodes closest and second-closest to the topic´s coordinate, respectively. The overlay automatically reroutes a topic´s messages to its secondary node should its primary node fail. Currently, SeDAX determines the coordinate of a topic by hashing its name. This kind of topic allocation is static, which can lead to unintended load imbalances. In this paper, we propose a topic delegation mechanism to make the assignment of topics to nodes dynamic. Our proposed mechanism is the only existing method to improve the flexibility and resource management of the SeDAX architecture so far. We define the load of SeDAX nodes and coordinates at different levels of resilience. On this basis, we develop distributed algorithms for load balancing. Simulations show that significant load imbalance can occur with static topic assignment and that the proposed algorithms achieve very good load balancing results.
Keywords :
computer network security; distributed algorithms; overlay networks; resource allocation; telecommunication network reliability; Delaunay-triangulated overlay network; SeDAX; data redundancy; distributed algorithms; geocoordinates; load balancing; message broker; overlay nodes; primary nodes; publish-subscribe information-centric networking architecture; resilient data forwarding; resource management; secondary nodes; static topic assignment; topic allocation; topic delegation mechanism; Computer architecture; Load management; Load modeling; Measurement; Overlay networks; Resilience; Resource management;
Conference_Titel :
Network Operations and Management Symposium (NOMS), 2014 IEEE
Conference_Location :
Krakow
DOI :
10.1109/NOMS.2014.6838254