Revenue-Based resource management on shared clouds for heterogenous bursty data streams

  • Authors:
  • Rafael Tolosana-Calasanz;José Ángel Bañares;Congduc Pham;Omer F. Rana

  • Affiliations:
  • Dpto. de Informática e Ingeniería de Sistemas, Universidad de Zaragoza, Spain;Dpto. de Informática e Ingeniería de Sistemas, Universidad de Zaragoza, Spain;LIUPPA Laboratory, University of Pau, France;School of Computer Science & Informatics, Cardiff University, United Kingdom

  • Venue:
  • GECON'12 Proceedings of the 9th international conference on Economics of Grids, Clouds, Systems, and Services
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

When data from multiple sources (sensors) are processed over a shared distributed computing infrastructure, it is necessary to often provide some Quality of Service (QoS) guarantees to each data stream. Service Level Agreements (SLAs) identify the cost that a user must pay to achieve the required QoS, and a penalty that must be paid to the user in case the QoS cannot be met. Assuming the maximisation of the revenue as the provider's objective, then it must decide which streams to accept for storage and analysis; and how many (computational / storage) resources to allocate to each stream in order to improve overall revenue. We propose an infrastructure for supporting QoS for concurrent data streams to be composed of self-regulating nodes. Each node features an envelope process to accept user streams; and a resource manager to enable resource allocation, admission control and selective SLA violations, while maximizing revenue.