High Performance Computing (HPC)

High Performance Computing (HPC)

Facility for processing research data and performing complex calculations at high speeds Facility
Facility for processing research data and performing complex calculations at high speeds
High performance computing (HPC) is all about scale and speed. HPC enables you to perform calculations much faster than can be achieved using a standard computer. HPC is a time-saving solution with the ability to make complex scientific research issues transparent. This can open up new opportunities, result in cost reduction, and improve research quality.

HPC enables you to solve computational problems for which a single computer lacks the power. An HPC cluster runs on Linux software, a computer operating system slightly less known than, for instance, Windows. A cluster consists of a collection of a huge number of individual computers called nodes that are connected to each other via a high-speed connection. All these different nodes in a cluster work in parallel with each other, therefore increasing the processing speed to deliver high-performance computing. Whether you work with human or non-human subject data, there is always a HPC cluster available that suits your research.

Depending on the type of cluster, it is possible to make use of bulk data storage, including backup facilities. This ensures that your data are always stored safely. Furthermore, several analysis pipelines are available and data can be shared securely and conveniently.

Which services do the UMCG HPC clusters offer?

  • They are able to solve computational problems for which a single PC lacks the power.
  • They provide temporary storage (tmp) for high performance computing.
  • They allow the use of bulk data storage (permanent storage, prm), including backup facilities.
  • They offer support with bioinformatics analyses.
  • Several analysis pipelines are available (for the analysis of DNA, RNA, and microbiomes, global screening array and analysis, and metagenomics).
  • A solid scheduler system is provided: calculation jobs are automatically scheduled and performed when it is your turn and capacity has become available.
  • Users can be granted certain rights or be provided with specific roles.
  • Data sharing is possible.

Please note: Knowledge of Linux software is required to work with the UMCG HPC cluster because you work with a command line.

  • Gearshift is the High Performance Computing (HPC) cluster for the UMCG’s human subject data. This cluster allows you to analyse data sets containing large and/or complex data and to perform calculations that cannot be achieved using a regular computer, in compliance with the rules and regulations regarding human subject data. The Gearshift cluster works with a scheduler system, which means that calculation jobs are automatically scheduled and performed when it is your turn and capacity has become available.

    The key features of the Gearshift cluster

    • Linux OS: CentOS 7.x with Spacewalk for package distribution and management;
    • Fully virtualized on an OpenStack cloud;
    • Deployment of HPC cluster with Ansible playbooks, under version control using a Git repository: league of robots;
    • Job scheduling: Slurm Workload Manager;
    • Account management: local administrator users and groups are provided with Ansible playbooks;
    • Regular users and groups use a dedicated Lightweight Directory Access Protocol (LDAP) for this cluster and are provided with either Ansible playbook or information from federated Algorithmic Aspects in Information and Management (AAIM);
    • Module system: Lmod;
    • Deployment of software, including bioinformatics software, using EasyBuild.

    Cluster Components
    Gearshift consists of various types of servers and storage systems, some of which can be directly accessed by users.

    Costs
    The costs you will have to pay for using the data storage service and the Gearshift HPC cluster are based on the amount of data you store. If you wish to use the data storage service, you will have to pay for the permanent storage (prm) option. Automatic tape backups are included in the price. If you wish to make HPC calculations, you will have to pay for the temporary storage (tmp) option as well, which includes use of the computing facilities. For both the permanent and the temporary storage services, the fee is €250 per terabyte (TB) of data storage per year. Therefore, if you wish to use both the storage and computing facilities, you will have to pay €500 per TB per year. The minimum amount required is 1 TB for 1 year.

    Training
    Beginner courses Human Subject Data

    Provided by
    The Genomics Coordination Center (GCC)

    For whom
    UMCG and UG researchers and/or research initiated by UMCG/UG researchers

  • The Peregrine cluster is a High Performance Computing (HPC) cluster that allows you to analyse data sets containing large and/or complex data. It allows you to perform calculations that cannot be achieved using a regular computer, in compliance with the rules and regulations that apply to non-human subject data.

    Please note: This HPC cluster is provided by the University of Groningen (UG). For more information about this cluster, please contact the Center for Information Technology service desk at the UG.

    Which services does the Peregrine HPC cluster offer?

    • It is able to solve computational problems for which a single PC lacks the power.
    • It provides three variants with different numbers of nodes (the available internal disk space per node is 1 TB).
    • The cluster has 463 TB of disk space.
    • It is suited to tasks that use more than one machine.

    Costs
    None

    Training
    You are welcome to participate in the monthly course to learn the basics needed to use the Peregrine cluster.
    More information and request form Non-Human Subject Data

    Provided by
    The University of Groningen (UG)

Before you can start using a UMCG HPC cluster, you are required to

  • Have successfully completed the associated cluster course or have had an intake interview with the Genomics Coordination Center (GCC);
  • Have been approved by the owner of a group to get access to that group and the associated data set you have to create a data set yourself.
  • The use of one of the HPC clusters requires knowledge of bioinformatics and how to work with a command line. To gain access to the cluster’s HPC facilities, you must have successfully completed the associated cluster course or have had an intake interview with the GCC. Depending on the number of applications, an introduction course is provided together with the Center for Information Technology (CIT) at the University of Groningen (UG) twice a year. The scheduling of the courses will take place based on demand
    If you are interested: signing up for Cluster course

  • Working with the HPC clusters takes place in a group setting. Gaining access to a group and the associated data set requires the prior approval by the owner of that group. Principal investigators can ask for a new group to be set up for their research project(s). Each group has at least one owner, one data manager and one member.

Contact

University Medical Center Groningen (UMCG)
Department of Genetics - Genomics Coordination Center
PO Box 30.001
9700 RB Groningen
The Netherlands

Visiting address
University Medical Center Groningen (UMCG)
Department of Genetics - Genomics Coordination Center
Antonius Deusinglaan 1
9713 AV Groningen

location: building 3211 and building 3226 (5th floor)