Nowadays there are many devices that generate data continuously for instance data from sensors at buildings, cars, energy meters and other devices. This data must be processed close to where it is produced for quick reaction to events, privacy issues and also to avoid sending data through the network to be processed in somewhere else (e.g., the cloud). The computing devices that process this data are at the edge and typically have low resources (memory, type of CPUs, bandwidth ..) and therefore, limited processing capabilities. In this paper we summarize the main features of two data stores, KeyDB and MariaDB, that can run on a Raspberry Pi. KeyDB is an in-memory key-value data store, while MariaDB is a relational database. We use the multi-core features of both data stores and also in a distributed scenario for KeyDB to evaluate and compare their performance using YCSB benchmark.