Scheduling multiple containerized tasks to edge computing nodes requires toolchains that lay out the necessary computational/communication models, algorithms, and methodology. The proposed toolchain performs stochastic optimization using simulated annealing (SA) and aims to support automated pod assignment to edge computing nodes. Different input parameters specify pod computational and communication requirements and edge nodes performance characteristics, as well as mapping options and algorithmic characteristics that affect the quality of the solution. Preliminary results from an eHealth-related digital twin (DT) use-case demonstrate limited load imbalance when scheduling DT workloads consisting of mixed HTTP PUT/POST and GET workloads to worker nodes (Raspberry Pi4) of a Kubernetes edge computing cluster. DT requests support basic create, read, update, and delete (CRUD) operations for modeling virtual real-time replicas of ECG biosensors within a Linux server-based open-source Eclipse Ditto DT management system that integrates the MongoDB database. Finally, we examine database server performance for simple DT-like CRUD operations on a Linux server. Results indicate that LittleD, a lightweight, low-cost, limited SQL-based database, designed initially for small systems, offers a much smaller average latency than MongoDB (used in Eclipse Ditto), InfluxDB, or MariaDB.