5. Conclusion and Future Development
Metagenomic studies are becoming increasingly widespread, yielding important insights into microbial communities covering diverse environments from terrestrial to aquatic ecosystems. With the advent of high-throughput sequencing platforms, the use of large scale shotgun sequencing approaches is now commonplace.
In a previous work we discussed an architecture and the performance of a prototype based on low-power Systems-On-Chip for metagenomic analysis able to support a fixed number of routinely analysis per day. In this paper we presented an evolution of such architecture, which supports the possibility to dynamically increase or decrease the sampling rate when critical situations occur. We analyzed four different strategies and we concluded that, while the previous architecture is an effective solution when a single analysis per device is performed every day, the best solution when the frequency increases - considering both cost and performance - is to “move” computational services from the Edge to the Fog or Cloud infrastructures, depending on the available Internet connection.
