Implement a function to traverse cell dependencies in an Excel-like spreadsheet and detect circular references between cells.

Focus on designing an efficient algorithm to traverse the cell dependencies, likely using depth-first search (DFS) or breadth-first search (BFS). Be prepared to discuss how to represent the spreadsheet's cells and their dependencies, possibly using a graph structure. Pay special attention to how you will track visited nodes to detect cycles, and consider edge cases such as empty cells or cells with no dependencies. Additionally, explain how your solution handles large datasets and what optimizations you might implement to improve performance.