Scientific Computing and Research Support Unit


 

dcsr_page.png

Explore our new offer on high performance computing and storage facilities

Request for resources

Requests for resources, whether consisting of storage and/or computing resources, can be done via the Catalogue de services of the Centre informatique or directly through one of these links:

Request for research data storage

Request for computational resources

Request for virtual machines (workstations or servers)

Operational and unit costs (UNIL authentication)

Costs calculator

Mission

The mission of the Scientific Computing and Research Support Unit  (Division calcul et soutien à la recherche, DCSR, in French) is to provide computation and storage resources to the UNIL community as well as transverse expertise to the various faculties of the University and its affiliated institutes. The expertise ranges from high performance computing (HPC) programming support to full stack development consulting missions.

DCSR Structure

The DCSR consists of technical and sysadmin specialists seconded by a team of scientific advisors/consultants. The technical and sysadmin specialists are in charge of maintaining, operating and evolving the infrastructure stack in order to provide:

  • computing resources over non-sensitive and sensitive data,
  • storage resources for non-sensitive and sensitive data.

The scientific advisors act as consultants to the faculties and are engaged on specific research projects to support and help bootstrap researchers on the usage of the available, computing and storage resources. They also offer consulting services on scientific, statistical, storage and web development projects that researchers may have.

Computing and Storage Infrastructure available in the DCSR

The infrastructure is set up to cater for a wide variety of storage needs: high performance storage for an HPC infrastructure, encrypted storage for the management of personal and/or sensitive data, various levels of data redundancy depending on the criticality of the data to be stored.  

The division maintains:

  • an HPC infrastructure,
  • a storage infrastructure supporting different types of data operating or not in an HPC environment,
  • a secured infrastructure for sensitive data (stay tuned, not yet launched)
  • a virtual infrastructure (workstations and servers)

and provides in addition support to researchers in terms of HPC, guidance in predictive analytics, machine learning, big data, as well as full stack development (web and database support).

Access to Resources

To access storage or computing resources, users must be part of a research group. Each research group is headed by a principal investigator (PI) who maintains the users in his group.

The PI must be legitimized by the faculty/institute he/she is attached to in his role of principal investigator ; this allows him to engage the expenses inherent to the usage of the HPC or storage infrastructure. Note that the PI is not necessarily required to have full control of a budget and/or to be able to manage or hire research staff.

The DCSR team

Manager: Roberto Fabbretti PhD

Systems Support:

HPC: Ewan Roche PhD

Storage: Roberto Fabbretti PhD

Virtualization and system administration: Marco Barroso

Applicative Support: 

HPC: Emmanuel Jeanvoine PhD

Data Science, Machine Learning: Philippe Jacquet PhD

Web, Databases: Thierry Lombardot PhD

Data valorisation:

Data modeling: Marion Rivoal PhD

Web, Databases: Loic Jaouen

Ticketing: Help desk Ci-UNIL

Training courses: due to the COVID-19, the next courses will be given remotely.

Course Date

Research - Introduction to Linux (commands and tools) for HPC

Tuesday 7th September - from 9 to 12am

Research - Introduction to reproducible research using version control

Tuesday 7th September - from 2 to 5pm

Research - Introduction to using the HPC clusters

Thursday 9th September - from 9 to 12am and from 2 to 5pm

Research - A Gentle Introduction to Deep Learning with Python and R

Beginning of 2022

Research - A Gentle Introduction to Decision Tree and Random Forest with Python and R

Thursday 21th October - from 9 to 12am

Thursday 25th November - from 9 to 12am

 

Once the course is announced you can register here: https://conference.unil.ch/cours/ci/

Contact DCSR

Roberto Fabbretti PhD
Roberto.Fabbretti@unil.ch
+41 21 692 40 32

Terms and Conditions

 Terms and Conditions

Partagez: