Power Aware SQL

Abstract

Database management systems(DBMS) comprise various algorithms for efficiently retrieving and managing data. Typically, algorithm efficiency or performance is correlated
with execution speed. However, the uptime of battery powered mobile and embedded systems strongly depends on the energy consumption of the involved components. This project reports our results concerning the energy consumption of different implementations of select algorithms on different types of queries (operations). We demonstrate that high performance algorithms often require more energy than slower ones. We also calculated different cost functions which are the measure of disk block access during query execution and also one of the biggest sources of energy dissipation. The reasonable energy consumption are comprised of seek energy, rotational kinetic energy due latency, due to block transfer and the main the stream is disk block accesses. Standard parameters of Seagate hard disks that is power and performance values are used to compute power consumption