Software-Defined Networking is a network framework that involves in separating a network’s control functions from its data forwarding functions, centralizing its intelligence, and abstraction its underlying architecture from applications and services. Hybrid Software-Define Networking is a framework which supports both the legacy protocols and the Software Defined Networking protocols to operate in the same environment.  In this paper, we present an efficient technique for finding the shortest path between any two nodes in Hybrid SDN by reducing the complicated network topology to a basic topology while taking hop count into account. In the suggested proposal, a subgraph is generated from a complicated network, and the best path is found using a modified Bellman-Ford method, which saves time. The emulation test bed in Mininet was established for various numbers of nodes, and the results were analysed. The modified Bellman-Ford algorithm selects the optimal path between the two nodes. For networks with different numbers of nodes, the proposed approach offered a significant increase in the TCP throughput when compared with that of Spanning Tree Protocol. The number of packets received by the SDN controller is less with the proposed work. This increases TCP throughput by reducing the broadcast packets.