Cloud computing apparently helps in reducing costs and providing the scheduling optimal level. In practice however it may confront the problem of unavailability of resources. Taking into consideration the cloud computing bank with its somehow commercial nature, the resources unavailability, such as liquidity risk, remains. In this paper, an attempt to show through a solution so far applied in economy, how would it be possible to predict such a liquidity risk in cloud computing bank. The proposed solution can especially be adapted to stock management. To reduce the risk we will also make use of a method inspired from physics based on the fluids mechanics; it is an application of Bernoulli's theorem called Torricelli. The resource bank will be considered as a reservoir of liquid, and the availability of resources then will depend on the liquid flow velocity and the replacement.