We put forward an algorithm on friend-recommendation of social networking sites based on SimRank and ant colony optimization, which broadens the appliance of the algorithm in this academic question. The algorithm focuses on the existing relationships between the members as the initial measurement and constructs artificial ants completed routing graph. Finally, an ordered and limited list of personalized recommendations through recursive optimization is produced. In the end, we verify the algorithm s rationality and validity through simulation and the result shows that it can improve the precision of friend-recommendation.