Snooze: A Self-Organizing, Self-Healing and Self-Optimizing Cloud Management System

Eugen Feller 1 Christine Morin 1 Matthieu Simonin 1 Anne-Cécile Orgerie 1 Yvon Jégou 1
1 MYRIADS - Design and Implementation of Autonomous Distributed Systems
IRISA-D1 - SYSTÈMES LARGE ÉCHELLE, Inria Rennes – Bretagne Atlantique
Abstract : Large-scale virtualized data centers enabling today's cloud services now require the cloud providers to implement easily configurable, fault-tolerant, and energy-efficient cloud management systems. In this demo we present the self-organization, self-healing, and self-optimization mechanisms of Snooze, a novel cloud management system based on a hierarchical architecture. Particularly, we focus on four key features: (1) dynamic hierarchy construction; (2) hierarchy failure recovery; (3) periodic and event-based (e.g. underload) virtual machine live migration for idle time creation; (4) power management to transition idle servers into a low-power state for energy savings.
Complete list of metadatas

https://hal.inria.fr/hal-00925656
Contributor : Eugen Feller <>
Submitted on : Wednesday, January 8, 2014 - 1:47:40 PM
Last modification on : Thursday, November 15, 2018 - 11:57:44 AM

Identifiers

  • HAL Id : hal-00925656, version 1

Citation

Eugen Feller, Christine Morin, Matthieu Simonin, Anne-Cécile Orgerie, Yvon Jégou. Snooze: A Self-Organizing, Self-Healing and Self-Optimizing Cloud Management System. The 10th USENIX International Conference on Autonomic Computing (ICAC 2013), Jun 2013, San Jose, CA, United States. ⟨hal-00925656⟩

Share

Metrics

Record views

702