Abstract: In order to achieve the high reliability of real-time systems in virtualized clouds, a fault-tolerant scheduling algorithm for virtualized clouds named FSVC that used primary-backup approach to tolerate physical host failures was proposed. FSVC strived to enhance the performance by employing backup-backup overlapping technique and virtual Machine migration technique. The constraints of the two techniques were elaborately analyzed to realize fault tolerance. Besides, a two-phase policy was incorporated in FSVC to further improve the performance. Simulation experiments demonstrated that FSVC can improve the schedulability and resource utilization effectively in virtualized clouds.