Autonomic network management is an innovative vision promising new horizons of efficient networking systems free from human control. This promise has, thus far, ushered in enormous yet dispersed research contributions in both industry and academia. The work presented in this article aims at putting these efforts into perspective deriving a more holistic view of the literature in this area. We analyze the requirements and the main contributions for the building blocks of any autonomic network management system (ANMS). We then describe a coherent classification methodology to compare existing ANMS architectures. Based on this analysis, we suggest a reference framework and highlight some open challenges and describe new research opportunities.

/pdf/towards-autonomic-network-management-an-analysis-of-current-70la1z72vy.pdf

Towards Autonomic Network Management: an Analysis of Current and Future Research Directions

Cloud computing requires a robust, scalable, and high-performance infrastructure. To provide a reliable and dependable cloud computing platform, it is necessary to build a self-diagnosis and self-healing system against various failures or downgrades. This paper is the first to study the self-healing function, a challenging topic in today's clouding computing systems, from the consequence-oriented point of view. To fulfill the self-diagnosis and self-healing requirements of efficiency, accuracy, and learning ability, a hybrid tool that takes advantages from Multivariate Decision Diagram and Naive Bayes Classifier is proposed. An example is used to demonstrate that this proposed approach is effective.

Self-healing and Hybrid Diagnosis in Cloud Computing

Les administrateurs, les developpeurs, les concepteurs logiciels ont besoin de nouvelles approches pour faire face a la complexite croissante des systemes informatiques. Afin de liberer les administrateurs de tâches repetitives et d'ameliorer la reactivite des systemes, ces systemes, alors appeles systemes autonomiques, tendent a etre dotes de capacites d'auto-gestion telles que l'auto-configuration et l'auto-reparation. Ces capacites d'auto-gestion sont fortement couplees a la capacite du systeme a s'auto-adapter, c.-a-d. a modifier sa structure et son comportement alors qu'il est en cours d'execution. Les approches se sont jusqu'alors concentrees sur l'auto-adaptation structurelle des systemes a base de composants logiciels en remaniant les liaisons entre les composants. Dans notre approche, nous nous focalisons sur l'adaptation comportementale. Nous avons defini le modele de composants MOCAS (Model Of Components for Adaptive Systems) pour permettre la realisation de systemes autonomiques a base de composants logiciels auto-adaptatifs. MOCAS repose sur une approche uniforme exploitant l'ingenierie des modeles : les capacites d'adaptation et les proprietes autonomiques de chaque composant reposent sur l'usage qui est fait par MOCAS du langage de modelisation UML. La structure d'un composant MOCAS repose sur les elements natifs du langage MOCAS. Le comportement du composant est decrit avec une machine a etats MOCAS. Cette machine est ensuite executee par le composant pour realiser son comportement. Pour devenir adaptable, le composant est deploye dans un conteneur respectant ce meme modele. Une boucle de controle a base de composants MOCAS permet de realiser des composants auto-adaptables de facon decentralisee et de les doter de proprietes autonomiques. Les politiques autonomiques exploitent alors les modeles de machines a etats afin d'auto-configurer et d'auto-reparer les composants.

https://tel.archives-ouvertes.fr/tel-00472005/document

MOCAS : un modèle de composants basé états pour l'auto-adaptation

NASA ANTS autonomous nano-technology swarm missions will be operating in the universe, and therefore rely much on high autonomy. This paper presents a novel technology for NASA's ANTS missions, named as model-driven autonomic computing. As the foundation for the technology, a new model is constructed for the ANTS system. Exceeding other existent models, the new hierarchical model overcomes the challenges of largeness, complexity, dynamicity and unexpectedness possessed by the ANTS system. Then, the paper exhibits the structure and functions of virtual neuron that is basic unit together with the model for the model-driven autonomic technology in ANTS missions. The paper also deploys self-configuration, self-healing, self-optimization and self-protection for ANTS. A case study, examples and simulations are illustrated.

Modeling for NASA Autonomous Nano-Technology Swarm Missions and Model-Driven Autonomic Computing

A swarm robotics system is a special type of wide-area and large-scale distributed system, which focuses on a group of robots cooperating to achieve the same goal using swarm intelligence. To treat the swarm robotics system with the self-healing and self-reproduction functions, this paper studied a prototype model based on the virtual neurons, autonomous self-diagnosis, consequence-oriented prescription, autonomous self-curing, and self-reproduction. This prototype system with self-healing has been implemented in the Trusted Electronics and Grid Obfuscation (TEGO) research center. Several practical cases were studied to show the effectiveness and efficiency of the model. The results demonstrate that the self-healing mechanism makes the system more reliable and the performance much improved, not only against failures, but also against failure propagations

A Prototype Model for Self-Healing and Self-Reproduction In Swarm Robotics System

The rapidly increasing complexity of computing systems is driving the movement towards autonomic systems that are capable of managing themselves without the need for human intervention. Without autonomic technologies, many conventional systems suffer reliability degradation and compromised security. Autonomic management techniques reverse this trend. This study describes the roles and functions of various autonomic components, and systematically reviews past and current approaches that have been/are being developed to address specific areas of the autonomic computing environment that focus on improving system dependability including both reliability and security concerns. Analyzing past research can lead to the design of a more advanced, dependable autonomic computing system. A novel and promising prototypical system that is a work-in-progress will be presented finally.

Tom Marshall

Papers

Autonomic and Dependable Computing: Moving Towards a Model-Driven Approach