Media content over the Internet has massively been increasing, resulting in popularity gain of cloud computing. Cloud computing is the only solution in hand to handle rapidly increasing digital media content. Through cloud computing, digital media can be manipulated, stored, and communicated in a much better and easier way. But due to increase in user’s demands and diversity of applications, it is, at times, not possible for a single cloud to fulfill all the requests. At that point, multiple clouds have to communicate and share resources through an intermediary, called cloud broker. To handle requests properly, broker has to predict the amount of resources required when a service is requested. This paper focuses on this particular issue. We present resource prediction part of our model here, along with its implementation and evaluation using CloudSim toolkit.