Postdoc position at Orange Lab

Post doctoral work description:

General

  • Type of contract : post-doctoral position for a duration of 12 to 18 months
  • Start Date : avril/mai 2011
  • Domain : cloud computing, autonomic management, feedback control,
  • Organization : France Télécom, R&D Division, laboratoire MAPS/MEP,
    38 – 40 avenue Général Leclerc
    98242Issy Les Moulineaux cedex, FRANCE
  • Locality : Issy Les Moulineaux, France
  • Supervisor : R Krishnaswamy, ruby.krishnaswamy@orange-ftgroup.com, Nicolas Rivierre, nicolas.rivierre@orange-ftgroup.com

Context

The emergence of cloud computing, self-service, on-demand and pay-per-use model, hastens the need for intelligent management and decision making. While technology innovations are clearly mandatory (interoperability, security, compound service provisioning, performance, scalability …), successful providers will provide the best services while incurring the least cost for themselves. The objective is to put in place adaptive resource management; adapting to variations in application workload, multi-tenancy, technology evolutions etc. The scope of the work is to design and develop decision support for a set of use cases listed further on. To be successful, this work should conjointly provide:

  • Software architecture of decision making framework to develop management controllers for cloud assets, where assets may be applications or cloud resources such as compute, storage, network or energy.
  • Decision support tools (methodology, models, and algorithms) to optimally provision resources to applications in a dynamic cloud environment.

The proof of concept will be demonstrated by applying the developed framework and techniques to a set of use cases described within the following sections.

Problem domain

1) Storage services deployed on distributed and multi-site cloud storage

Applications offering content management services (e.g. mail servers, multi-media content servers, share-point services, file servers …) are managed entities structured as a set of interconnected VMs. Performance is conditioned by the quality of data and storage services with applications access the data store by different types of interfaces: file system, block or object.

2) Deployment and life-cycle management of multi-tier applications across multiple data centers

Management decisions must be taken either on arrival of new requests (new applications to deploy) or in the background such as at scheduled reconfigurations or for application work load changes.

Scope of work

Intelligent decision support systems analyse the environment, identify problems and propose corrective actions. Frameworks may be designed based on a paradigm of feedback-based control relying on policy or rule-based systems, where rules may be dynamically configured using on-line decision support tools.

We are opening a post doctoral position to build decision support frameworks to assist in autonomic management of data centres resources and applications deployed within data centres. This work will be conducted in liaison with experts in operations research.

The work will be organised as follows:

  • Study a complex service system such as a federated cloud storage system, so as to derive objectives and conditions for management.

  • In this study, specific optimisation and management problems will be identified, either in the perspective of consumers using a federated cloud storage system or in the perspective of the cloud storage provider.
  • Define design methodologies, including framework, tools for building a decision support system to address the management problems identified in the selected case study.

Collaborations

This work will be conducted jointly by two laboratories of France Télécom R&D. One laboratory focuses on middleware and platforms and the second on applied mathematics and operational research to develop tools for decision making.

Qualification

We are searching for a candidate with demonstrated research abilities and who is motivated to developing software. Applicants should have or soon complete a PhD degree in computer science in a subject relevant for the described area of work. The work will involve the elaboration of models, architecture, specification and development of software components.

The ideal candidate will be autonomous, creative and ambitious.

Environment

The candidate will work within the France Télécom R&D MAPS/MEP specialized in architecture and middleware platforms. The post doctoral work will be executed at Issy les Moulineaux. This work will be executed in collaboration with researchers in the domain of operations research.

Leave a Reply

 

 

 

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Rejoindre ResCom

Pour s'inscrire à ResCom il vous suffit d'envoyer un mail de demande d’adhésion à l'adresse sympa@cines.fr avec dans l'objet de votre e-mail subscribe asr.rescom Prénom Nom.

Vous pouvez aussi joindre le comité scientifique pour toute demandes/questions à partir de la page contact.

IL est aussi conseillé de s'inscrire à la liste di GDR asr-forum.

Pour plus d'information sur les listes, regardez la page "contacts".

Une aide sur sympa est aussi disponible.