Design and Implementation of OpenFlow Networks for Medical Information Systems

In current medical information networks, there is a problem because the cost of core switches is increasing owing to the need to meet demands in terms of throughput and robustness. To address this problem, we propose MediFlow as a medical information network based on OpenFlow. MediFlow replaces expensive, high-performance core switches with a MediFlow core network comprising multiple inexpensive and fully connected OpenFlow switches. The MediFlow core network adaptively allocates the communication capacity to time-varying traffic in a medical information network. If a link fails, MediFlow dynamically assigns a path to ensure that the communication system remains robust. We implemented MediFlow using commercial OpenFlow switches. The experimental evaluation showed that MediFlow could improve the throughput by allocating individual paths to each flow about 7 s after the occurrence of congestion by converging multiple flows. In addition, even if a link failure occurred in the path used by the flow, MediFlow re-established a path with a disconnection time of about 0.7 s.

[1]  A. Murat Tekalp,et al.  Scalable video streaming over OpenFlow networks: An optimization framework for QoS routing , 2011, 2011 18th IEEE International Conference on Image Processing.

[2]  Nick McKeown,et al.  OpenFlow: enabling innovation in campus networks , 2008, CCRV.

[3]  A. Murat Tekalp,et al.  A distributed QoS routing architecture for scalable video streaming over multi-domain OpenFlow networks , 2012, 2012 19th IEEE International Conference on Image Processing.

[4]  Haitao Wu,et al.  Explicit Path Control in Commodity Data Centers: Design and Applications , 2016, IEEE/ACM Transactions on Networking.

[5]  Keith McCloghrie,et al.  Structure and identification of management information for TCP/IP-based internets , 1988, RFC.

[6]  Marshall T. Rose,et al.  Management Information Base for network management of TCP/IP-based internets , 1990, RFC.

[7]  Filip De Turck,et al.  Optimizing scalable video delivery through OpenFlow layer-based routing , 2014, 2014 IEEE Network Operations and Management Symposium (NOMS).

[8]  Keith McCloghrie,et al.  Structure and identification of management information for TCP/IP-based internets , 1988, RFC.

[9]  Yu Cao,et al.  Explicit multipath congestion control for data center networks , 2013, CoNEXT.

[10]  Kohei Ichikawa,et al.  A simple multipath OpenFlow controller using topology‐based algorithm for multipath TCP , 2015 .

[11]  Deng Pan,et al.  OpenFlow based Load Balancing for Fat-Tree Networks with Multipath Support , 2013 .

[12]  Zdravko Bozakov,et al.  Flow-based load balancing in multipathed layer-2 networks using OpenFlow and multipath-TCP , 2014, HotSDN.

[13]  Byrav Ramamurthy,et al.  Network Innovation using OpenFlow: A Survey , 2014, IEEE Communications Surveys & Tutorials.

[14]  Jeffrey D. Case,et al.  Simple Network Management Protocol (SNMP) , 1989, RFC.

[15]  Mark Handley,et al.  TCP Extensions for Multipath Operation with Multiple Addresses , 2020, RFC.

[16]  Michael Karl,et al.  Multimedia optimized routing in OpenFlow networks , 2013, 2013 19th IEEE International Conference on Networks (ICON).

[17]  Qi Hao,et al.  A Survey on Software-Defined Network and OpenFlow: From Concept to Implementation , 2014, IEEE Communications Surveys & Tutorials.

[18]  Acee Lindem,et al.  Virtual Router Redundancy Protocol , 1998, RFC.

[19]  Mark Handley,et al.  Improving datacenter performance and robustness with multipath TCP , 2011, SIGCOMM 2011.