Remote attestation is an important mechanism to provide the trustworthiness proof of a computing system by verifying its integrity. In this paper, we propose an innovative remote attestation framework called DR@FT for efficiently measuring a target system based on an information flow-based integrity model. With this model, the high integrity processes of a system are first verified through measurements and these processes are then protected from accesses initiated by low integrity processes. Also, our framework verifies the latest state changes in a dynamic system instead of considering the entire system information. In addition, we adopt a graph-based method to represent integrity violations with a ranked violation graph, which supports intuitive reasoning of attestation results. Our experiments and performance evaluation demonstrate the feasibility and practicality of DR@FT.