DSpace logo

Please use this identifier to cite or link to this item: http://dspace.bits-pilani.ac.in:8080/jspui/xmlui/handle/123456789/8244
Full metadata record
DC FieldValueLanguage
dc.contributor.authorHaribabu, K-
dc.date.accessioned2023-01-03T07:08:56Z-
dc.date.available2023-01-03T07:08:56Z-
dc.date.issued2019-
dc.identifier.urihttps://ieeexplore.ieee.org/document/8711070-
dc.identifier.urihttp://dspace.bits-pilani.ac.in:8080/xmlui/handle/123456789/8244-
dc.description.abstractCapturing and monitoring the global state of the network in a software defined network (SDN) is crucial for efficient routing, performance monitoring, Quality of Service (QoS) assurance, etc. The two major existing approaches for statistics collection in SDN are polling-based and event-based. Due to the asynchronous nature of the network, statistics collected through polling have inconsistencies and are not suitable for capturing the consistent global state of the network. On the other hand, event-based monitoring schemes may give sparse information about the network. Globally consistent state detection is well studied for asynchronous systems. However, current SDN standards such as OpenFlow do not support any functionality to collect globally consistent statistics. In this paper, we propose “OpenSnap”, an algorithm to determine the globally consistent state of the system. To support OpenSnap, we extend the OpenFlow protocol by adding a new action. The experimental results show that the statistics collected at the SDN controller using the proposed OpenSnap algorithm are always consistent.en_US
dc.language.isoenen_US
dc.publisherIEEEen_US
dc.subjectComputer Scienceen_US
dc.subjectSDNen_US
dc.subjectOpenFlowen_US
dc.subjectNetwork monitoringen_US
dc.subjectConsistencyen_US
dc.titleOpenSnap: Collection of Globally Consistent Statistics in Software Defined Networken_US
dc.typeArticleen_US
Appears in Collections:Department of Computer Science and Information Systems

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.