This paper focuses on minimizing energy consumption by addressing the joint microservice (MS) placement and computing resource allocation problem in distributed MS-aware wireless cellular networks (DMS-WCNs). We propose a paradigm in which each large service is composed of several lightweight MSs distributed among different small base stations (SBSs) to perform individual functions. For an arbitrary service request, the macro base station (MBS) invokes the SBSs that have cached the necessary MSs to execute the corresponding computational tasks. Once the computation is completed, the SBSs send the results back to the MBS, which then integrates and delivers the final result to the user. Taking into account the practical considerations of users’ service latency requirements and SBSs’ limited caching and computing resources, we formulate the minimization problem. To solve it efficiently, we develop a two-stage approach. In the first stage, we derive the closed-form expression of the computing resource allocation policy with regard to the MS placement. In the second stage, we introduce the swapping-oriented algorithm to explore an improved MS placement strategy. The simulation results demonstrate that our proposed algorithm achieves close-to-optimal performance compared to the exhaustive algorithm and significantly outperforms the other benchmark strategies.