Hardware leverages for energy reduction in large scale distributed systems
Résumé
This chapter provides the summary and references on existing and future leverages to adapt the underlying hardware infrastructures of large-scale distributed computing systems in order to decrease their energy consumption. It is intended to be used by researchers, as well as companies and general public, to drive their research fostering activities toward energy-aware solutions for large-scale distributed computing systems. The chapter addresses one hardware component, such as the central processing unit (CPU), main memory, storage (disk and flash), motherboard, network interface, and fans. Leverages are available at several levels, from such subsystems to changing the state of whole computers. The chapter focuses on the main features that impact the operation modes of the CPU and include three mechanisms, namely, thermal control zone, performance state (P-state), and CPU's state. It also focuses on dynamic random-access memory (DRAM), which is the common memory used in servers.