NPi-Cluster: A Low Power Energy-Proportional Computing Cluster Architecture | Zendy

Sebastiao Emidio Alves Filho | Zendy; Aquiles Medeiros Filgueira Burlamaqui | Zendy; Rafael Vidal Aroca | Zendy; Luiz Marcos Garcia Goncalves | Zendy

AI Assistant Blog Pricing

Home ZAIA Blog

Open Access

NPi-Cluster: A Low Power Energy-Proportional Computing Cluster Architecture

Author(s) -

Sebastiao Emidio Alves Filho,

Aquiles Medeiros Filgueira Burlamaqui,

Rafael Vidal Aroca,

Luiz Marcos Garcia Goncalves

Publication year - 2017

Publication title -

ieee access

Language(s) - English

Resource type - Journals

SCImago Journal Rank - 0.587

H-Index - 127

ISSN - 2169-3536

DOI - 10.1109/access.2017.2728720

Subject(s) - aerospace , bioengineering , communication, networking and broadcast technologies , components, circuits, devices and systems , computing and processing , engineered materials, dielectrics and plasmas , engineering profession , fields, waves and electromagnetics , general topics for engineers , geoscience , nuclear engineering , photonics and electrooptics , power, energy and industry applications , robotics and control systems , signal processing and analysis , transportation

This paper presents the NPi-Cluster, an energy proportional computing cluster that automatically powers ON or OFF the number of running machines according to the actual processing demand. A theoretical model is proposed, discussed, and implemented on a cluster composed of Raspberry Pi computer boards designed and built in order to test the proposed system architecture. Experimental results show adequate performance of the proposed platform when compared with other web servers running on traditional server architectures, but with considerably less power consumption. The power consumption of the entire cluster is about 14 W when running at maximum performance. In this situation, the system is able to handle more than 450 simultaneous requests, with about 1000 transactions per second, making it possible to be used as a server capable of handling real web workloads with acceptable quality of service. When the requests demand is reduced to a minimum, the power consumption is dynamically reduced until less than 2 W. Additionally, the proposed cluster architecture also provides high availability by reducing single points of failure on the system.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.

Having issues? You can contact us here

Accelerating Research