It is well-known that effective resource utilization is a critical factor in providing high quality microservicess in cloud computing. In a large-scale cluster, if each machine can save a small amount of resources, a huge effect could be made to significantly reduce the overall computing cost as the saved resources across the cluster can be gathered into a large resource pool to facilitate the computation as a whole. As such, how to effectively allocate the resources in a single host is critical to the success of this saving strategy. To this end, we propose a multi-dimensional resource allocation algorithm, called Hestia, for a single machine in a stand-alone environment with each dimension having limited resources. The algorithm is designed by leveraging dynamic programming (DP) techniques to squeeze the occupied resources of the existing microservices without compromising their performance, and leave the saved resources for other newly deployed microservices. Our experimental results show that compared with the default case, this method can save up to 15 of the resources for a single machine while ensuring the stability of online microservices.