Section: New Software and Platforms


Keywords: Energy management - Fault-tolerance - Self-organization - Self-healing - Cloud computing - Consolidation - Virtualization

Scientific Description

Snooze is a scalable, resilient and energy-aware virtual machine management framework for clouds. It is the result of Eugen Feller's PhD thesis which has been funded by the ANR EcoGrappe project (2008 - 2012).

The objectives of the Snooze ADT are threefold: (i) to distribute Snooze system as an open source software and to provide support to the user community (ii) to implement additional features to make it more user-friendly (iii) to integrate it with other open source software stacks and in public testbeds to favour its dissemination.

Snooze is a highly modular system for IaaS clouds. For the scientific community, it constitutes a unique framework for the experimentation of resource management policies in a real system. More generally, it allows any organization to operate a large-scale cluster as a resilient and energy-aware computing infrastructure enabling on demand provisioning of virtual clusters.

Functional Description

Snooze is a self-organizing and energy aware Cloud management framework.

Snooze is a novel Infrastructure-as-a-Service (IaaS) cloud-management system, which is designed to scale across many thousands of servers and virtual machines (VMs) while being easy to configure, highly available, and energy efficient. For scalability, Snooze performs distributed VM management based on a hierarchical architecture. To support ease of configuration and high availability Snooze implements self-configuring and self-healing features. Finally, for energy efficiency, Snooze integrates a holistic energy management approach via VM resource (i.e. CPU, memory, network) utilization monitoring, underload/overload detection and mitigation, VM consolidation (by implementing a modified version of the Sercon algorithm [59] ), and power management to transition idle servers into a power saving mode. Snooze is a highly modular piece of software. It has been extensively evaluated on the Grid'5000 testbed using realistic applications.

  • Participants: Eugen Feller, Christine Morin, Jiajun Cao, Gene Cooperman, Yvon Jégou, David Margery and Matthieu Simonin

  • Contact: Christine Morin

  • URL: http://snooze.inria.fr/