Design of an Edge Computing Testbed to Simplify Experimental Setup

Running IoT applications on edge computing infrastructures has the benefits of low response times and efficient bandwidth usage. System verification on a testbed is required to deploy IoT applications in production environments. In a testbed system, round-trip times (RTT) between servers and wireless base stations vary according to the locations of the servers and the wireless base stations in a network. When the server to place a Docker container running a test server program is determined, the RTT between the server and the wireless base station must be ensured to satisfy a target application’s response time requirement. However, existing testbed systems have no mechanism for ensuring the RTT between servers and wireless base stations. A testbed user must manually determine a server to place a Docker container. Furthermore, the determination involves a large amount of configuration data including RTTs between all pairs of servers and wireless base stations. In this paper, we present an edge computing testbed system that automatically determines which servers to place Docker containers on according to the testbed users’ preferred RTTs between the servers and the wireless base stations. The proposed system reduces the configuration data size to one divided by the number of servers.