Low Power Solutions: Autonomous Collaborative Low Power Computing Systems
This talk presents an overview of low power collaborative computing solutions for small scalable mini datacenter. Smart systems and infrastructure are a key challenge for delivering innovation in reliable low power computing systems. Needs of autonomous and adaptive solutions will guarantee challenges on having optimization on energy and computing efficiency.
This talk reports on applied research on the field of Low Power computing solutions in critical context as for example Polar Regions where necessity is on using Low Power computing systems and innovative low power radio-communication capabilities due to low energy constraints.
Prof. Tomoya Enokido
Rissho University, Japan
Green Computing: Issues, Algorithms and Solutions
In order to realize scalable, high performance and fault-tolerant distributed applications various types of computing systems like Cloud Computing, Peer-to-Peer (P2P) and Grid Computing systems are developed. For these systems, server cluster systems are widely used to provide applications with scalable, high performance and fault-tolerant services. A server cluster system is composed of a large number of servers where application processes are performed on multiple servers. Application services can be provided by balancing processing load among servers to perform application processes. But, a large amount of electric energy is consumed in a server cluster system since multiple servers consume electric energy to perform application processes. Hence, it is critical to discuss not only the realization of scalable, high performance and fault-tolerant application services, but also the energy-efficient server cluster systems in order to decrease the energy spending which leads to Green Computing.
In this talk, first I will introduce some issues on Green Computing and energy-saving systems. Then
based on experimental results, I will present server computation models and power consumption models to
perform different types of application processes. After that, I will present the load balancing algorithms to
select a server for each type of request process so that the total energy consumption of a server cluster and
response time of each process can be reduced. Finally, I will conclude the talk with some lessons learned
and will discuss challenges of Green Computing
