MANeUveR: MANagement agency for cloUd Resources

  • Director: Mădălina Eraşcu
  • Funding Agency: UEFISCDI (PN-III-P2-2.1-PED-2016-0550)
  • Project Period: September 2017 - December 2018
  • Host Institution: Institute e-Austria Timisoara

    • NEWS! There is a continuation of the research performed in MANeUveR which recently received a grant: SAGE: A Symbiosis of Satisfiability Checking, Graph Neural Networks and Symbolic Computation

    The financial support for MANeUveR has ended. The research activities related to the project continue.

    MANeUveR has won a mention award at BringITon2018, Iasi, December 13-14, 2018. BringITon2018 is a workshop for promoting and capitalizing the interaction between computer science in academia and business environment.

    Project Overview

    Cloud computing offers attractive options to migrate corporate applications without the software personnel (End Users - EUs) needing to manage any physical resources. While this "ease" is appealing, several issue arise:

    1. Which Cloud Providers (CPs) offer the best infrastructure at a fair budget?
    2. I am no Cloud expert then what are the characteristics of the infrastructure which best fit my application?
    To answer these questions one must solve a resource management problem, that is, the allocation of computing, storage, networking and (indirectly) cost resources to a set of applications such that the performance objectives of the application, CPs and EUs are jointly fulfilled. Efficient resource use is typically achieved through virtualization technologies, which facilitate statistical multiplexing of resources across the three parties. There are many approaches which answer separately these questions but there is no comprehensive and easily usable solution for these issues. MANeUveR solves this problem by integrating the following components:
    1. A Provider Acquisition Module, through a Web crawler, will periodically update a database with infrastructure details, in particular virtual machines (VMs) offers (CPUs number, memory, storage, price, operating system, number of IP addresses, transfer rate, etc.) from various CPs.
    2. An Application Description Module will provide the EU the capability to describe and store the VMs characteristics and application constraints in order to be consulted other times and to obtain a profile of the application.
    3. A Recommendation Module will provide a (sub)optimal solution for application deployment in the CP infrastructure regarding the number of VMs needed for deployment and their characteristics.
    Using a secure-billing e-mail service and secured web container applications, MANeUveR will demonstrate its effectiveness in real life. The tool will be based on an open-source core to motivate its wide adoption.

    Project Members

    Former Project Members

    Publications and Reports

    Github Repository

    https://github.com/Maneuver-PED




    Contact: Mădălina ERAŞCU.