Congratulations to Remo Andreoli for having received the Best Student Paper Award at the 11th International Conference on Cloud Computing and Services Science (CLOSER) 2021 with a paper about RT-MongoDB: a NoSQL database with differentiated performance, essentially his thesis work for the MSc in Computer Science, co-supervised by Tommaso Cucinotta and Dino Pedreschi. An extraordinary achievement for a student who just graduated :-)!
In the paper, a set of modifications to the popular MongoDB open-source NoSQL datastore are presented, which introduce a per-client/request prioritization mechanism within the request processing engine, enabling higher-priority requests to receive limited interference due to lower-priority ones. Extensive experimentation with synthetic stress workloads demonstrates that the proposed solution is able to assure differentiated per-client/request performance in a shared MongoDB instance. Namely, requests with higher priorities achieve reduced and significantly more stable response times, with respect to lower priorities ones. This constitutes a basic but fundamental brick in providing assured performance to distributed soft real-time cloud applications and services making use of NoSQL database services.
A representative plot from the paper is the one visible below, showing boxplots of the response times (on the Y axis) obtained in a given time window with several clients differentiated by priority levels (different boxplots), compared to the one obtained in the original unmodified MongoDB (rightmost boxplot). The plot refers to a 12-users scenario, with 4 users per priority level.
For more details, the paper is available here.