Abstract
Internet is expected to be available every time, so that the services run on it are able to be used any time. To achieve high availability Internet network, redundancy is used by employing more than one gateway so if one gateway is down, the other gateway will take the job of it. In the access layer, we use first-hop redundancy as the redundancy system. In this research, author design a system that has first-hop redundancy. The system is designed using software defined networking paradigm. The system uses POX as controller and Openflow as Controller Data-path Protocol Interface between switches and controller. The system is simulated on the mininet to examine the capability of the system by measuring some parameters: fail over delay, resource utilization and overhead on the bipartite topology. Of the examination results and analysis, the system is able to run the first-hop redundancy which yields fail-over delay below 140 ms for single flow.
Highlights
In the design of network, redundant link is mandatory to mitigate the effect when the primary link is down, in other word network redundancy
Bipartite topology is used with varied number L1 and L2 switch refers to the table 1
The 140 ms of the first-hop redundancy's fail-over delay is better than VRRP's 145 ms with 220 backup priority [3], so the performance is considered good
Summary
In the design of network, redundant link is mandatory to mitigate the effect when the primary link is down, in other word network redundancy. The availability of the network is expected to be 99.999% (five nines)[2]. Network redundancy can be implemented either in the core, distribution, or access infrastructure layer. There are some effort to get gateway redundancy in the Ethernet network or first-hop redundancy[3], such as Proxy Address Resolution Protocol[4], Virtual Router Redundancy Protocol[5], Hot Standby Router Protocol[6], and Common Address Redundancy Protocol[8]. Those protocol works in the conventional network control which is distributed system, every gateway maintain their own state and communicate with other devices periodically, so that the system is very complex. As more gateway employed in the first-hop redundancy, the system becomes more complex and greater overhead traffic
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call