With the advance of Location-Based Services (LBS), researches on trip recommendation have attracted extensive attentions. Among them, one active topic is trip planning. In the previous studies on trip planning, various user constraints such as travel time, travel budget, attraction categories, etc., have been considered and users' past travel logs were analyzed for travel recommendation. However, such kind of trip planning approaches cause the computational complexity to increase significantly. Hence, in this paper, we demonstrate a cloud-based travel recommendation system named TripCloud, which is built by extending our previous work, Personalized Trip Recommendation (PTR), for meeting user's multiple constraints with efficient trip planning. TripCloud encapsulates several data mining techniques and a cloud-based trip planning model to rate the interestingness of each attraction and plan an interesting trip, respectively. Visualization interface is also provided to exhibit the recommended trips based on the characteristics of user constraints.