LIBERO: a framework for autonomic management of multiple non-functional concerns

Authors:
Marco Aldinucci;Marco Danelutto;Peter Kilpatrick;Vamis Xhagjika
Affiliations:
University of Torino;University of Pisa;Queen's University Belfast;University of Pisa
Venue:
Euro-Par 2010 Proceedings of the 2010 conference on Parallel processing
Year:
2010

Citing 6
Cited 3

An Artificial Intelligence Perspective on Autonomic Computing Policies

POLICY '04 Proceedings of the Fifth IEEE International Workshop on Policies for Distributed Systems and Networks
Bringing skeletons out of the closet: a pragmatic manifesto for skeletal parallel programming

Parallel Computing
Behavioural Skeletons in GCM: Autonomic Management of Grid Components

PDP '08 Proceedings of the 16th Euromicro Conference on Parallel, Distributed and Network-Based Processing (PDP 2008)
Autonomic multi-agent management of power and performance in data centers

Proceedings of the 7th international joint conference on Autonomous agents and multiagent systems: industrial track
Resource-Definition Policies for Autonomic Computing

ICAS '09 Proceedings of the 2009 Fifth International Conference on Autonomic and Autonomous Systems
Autonomic management of non-functional concerns in distributed & parallel application programming

IPDPS '09 Proceedings of the 2009 IEEE International Symposium on Parallel&Distributed Processing

Model checking support for conflict resolution in multiple non-functional concern management

Euro-Par'11 Proceedings of the 2011 international conference on Parallel Processing
Coordinating multiple administration loops using discrete control

ACM SIGOPS Operating Systems Review
Coordinating self-sizing and self-repair managers for multi-tier systems

Future Generation Computer Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

We describe a lightweight prototype framework (LIBERO) designed for experimentation with behavioural skeletons--components implementing a well-known parallelism exploitation pattern and a rule-based autonomic manager taking care of some non-functional feature related to pattern computation. LIBERO supports multiple autonomic managers within the same behavioural skeleton, each taking care of a different non-functional concern. We introduce LIBERO-built on plain Java and JBoss-and discuss how multiple managers may be coordinated to achieve a common goal using a two-phase coordination protocol developed in earlier work. We present experimental results that demonstrate how the prototype may be used to investigate autonomic management of multiple, independent concerns.