The overall objective of the Phd is to investigate how HPC/Cloud distributed storage systems can integrate the notion of elasticity (i.e. vertical and horizontal scaling) in order to satisfy application requirements while minimizing "on-the-fly" resources costs.
Description
The overall objective of the Phd is to investigate how HPC/Cloud distributed storage systems can integrate the notion of elasticity (i.e. vertical and horizontal scaling) in order to satisfy application requirements while minimizing "on-the-fly" resources costs.
The work will be organized around two major periods:
First, the Phd student will (i) identify the advantages and the limitations of existing storage systems for HPC and Clouds (locally attached devices vs. remote storage solutions) and (ii) determine the pros and cons of providing the elasticity capability according to the envisioned use-cases.
Second, the PhD candidate will investigate how virtualization technologies and autonomous mechanisms can be used to deliver storage elasticity. Among the ways to be explored, we envisage to reorganize the file system deployment by using consolidation technics at the IaaS level or by using the possibility to attach/detach virtual storage systems on each virtual node. Finally, we want to offer a direct control to end-users so that they can decide how the distributed storage system should be deployed. Providing such an elasticity capability can be a mean to reduce the energy footprint of storage systems.
During this three-year position, the student will learn to design, implement and evaluate prototypes to validate their proposals. Such validations will include manipulating simulator toolkits and conducting large-scale in-vivo experiments on well-known testbeds such as Grid’5000 or even directly on public cloud platforms.
To apply, please email a cover letter, CV, contact address of at least two professional references and copies of degree certificates to Dr. Adrien Lebre Incomplete applications will not be considered or answered.
Nr of positions available : 1
Research Fields
Computer science
Career Stage
Early stage researcher or 0-4 yrs (Post graduate)
Research Profiles
First Stage Researcher (R1)
Benefits
The candidate will be mainly hosted in Nantes (France).
After the first year, the candidate will join the Storage Research Group of Barcelona Supercomputing Center under the supervision of Dr. Ramon Nou where he/she will finalize the study of the pros and cons of HPC and cloud storage systems and investigate the interest of adding the elasticity capabilities to existing storage solutions for the envisioned use-cases.
After the second year, the candidate will join the Dublin IBM Research Center under supervision of Dr. Bogdan Nicolae where he/she will investigate the pros and cons of low-level virtualization storage technologies (aka. vDisk).
Comment/web site for additional job details
bigstorage.oeg-upm.net/jobs.html
Requirements
| ENGLISH |
| Excellent |
| Computer science |
| Master Degree or equivalent |
| At the time of recruitment, the applicant must not have lived in France for more than 12 month in the previous 36 month and he/she is in the first four years of his/her research career, starting at the date of obtaining the degree which would formally entitle to embark on a doctorate. Master degree on computer systems (or similar) | |
| Knowledge on • Operating systems / system administration and development • Storage and (Distributed) File systems • Shell programming (Ruby or Python) • Object Oriented Programming (recent languages including functional abstractions such as Scala would be appreciated but are not mandatory) • Cloud Computing (Amazon Web Services, OpenStack, …) • Simulator toolkits Other skills • English (both spoken and written) • Open-minded and strong integration skills |