Path Selection With Critical Switch-Aware For Software Defined Networking

The flexibility of software defined network come at the cost of network rule configuration and load balancing during path selection. Network topology changes more often, similarly a high frequency of network failure was reported. The latter and former events cause devices to carry out rules update operations more frequently (inserting, deleting, and modifying rules). These operations are proportional to the number of rules in the device flowtable. The higher the flowtable utilization the higher the operation. Consequently, increases the chance of switch flowtable overflow. Therefore, upon occurrences of any of the two events, effective path computation is required to guide the routing of flows to the destination. However, the traditional path selection algorithms failed to consider the switch role during the path selection process. The key to address this challenge is to maintain the load balance among switches when selecting a path. Evenly network load distribution help to minimize the chance of network bottleneck in some part of the network. This paper proposed minimal Critical switch-based routing (CSBR) to select a path that can incur light rule update operation. CSBR approach is based on a simple and yet effective critical switch frequency score (CWFscore) metric which encourages path selection with a minimal number of critical switches. The experimental benchmark results show that the proposed algorithm has a better flowtable occupancy rate compared to the baseline work.