Graph Neural Networks (GNNs) have demonstrated exceptional performance in processing graph-structured data and have found widespread applications in domains such as recommendation systems and natural language processing. However, as graph sizes increase, training GNNs becomes computationally and memory intensive, necessitating effective distributed training approaches through graph partitioning. High-quality graph partitioning aims to minimize inter-subgraph connections while balancing computational and memory loads across partitions, thereby enhancing training efficiency. Existing partitioning methods are often constrained by evaluation metrics designed for graph analysis tasks, rendering them less effective for distributed GNN training scenarios. In this paper, we propose a novel Balanced Graph Partitioning Algorithm (BGPA), designed to achieve an optimal balance between computational and communication costs and to equalize memory utilization among computing nodes. We validate the effectiveness of our method by comparing its performance with multiple different partitioning methods on distributed GNN training using the Reddit and Ogbn-products datasets. Through extensive experimental evaluations, we demonstrate that our method significantly reduces the overhead associated with distributed GNN training, leading to substantial improvements in training efficiency. These results suggest that our approach offers a viable solution for scaling GNN training to large graphs, addressing a critical bottleneck in the application of GNNs to real-world, large-scale graph data.