Research Computing Platforms

We offer a range of advanced platforms for computationally intensive research. These services are free for UCL researchers to use on a fair share basis.

Overview

We currently provide two computing clusters: Myriad and Kathleen. Together, these platforms enable our users to perform all but the most demanding analyses and simulations. use the same Linux-based software stack, which provides a consistent user interface and makes it easy for researchers to switch platforms according to their need.

Kathleen

Kathleen was launched at the end of 2020 as a replacement for Grace, and is designed to take care of large parallel computing tasks. It is designed specifically for this purpose with a high speed Infiniband network linking together its compute nodes, allowing them to coordinate effort on a single task very effectively.

Myriad

Myriad was launched in 2018 as a solution for data-intensive workloads. It is designed specifically for this purpose with a high speed Infiniband network linking high performance storage with powerful compute nodes – ideal for working with large datasets.

Key features

CORE

• Computing power for demanding analyses and simulations.
• Different classes of architecture for different workloads.
• Expert advice and consultancy for complex research computing challenges.
• Specialist technical support available at rc-support@ucl.ac.uk
• Over 730 software applications. Packages can be installed by our staff (license permitting).
• Supported access to shared regional and national research computing facilities.
• Home directory (up to 150GB, depending on service) and backup.
• Operational review by academic governance group.

CORE+

• Purchase a range of dedicated compute nodes and storage that is hosted centrally.
• Options to manage access to nodes.
• Nodes typically installed within 4 weeks of the end of the quarter.

How to apply to use the research computing platforms

We provide access to two computing clusters: Myriad and Kathleen (What is a computing cluster?). All UCL researchers are eligible to use these platforms on a fair share basis and at no cost. We have a single application process for all of our research computing platforms - see the account services guide for details.

To obtain an account you must either be, or be sponsored by, a permanent member of staff. If you are a student or a postdoctoral researcher, this will normally be you supervisor.

You will need your UCL userid and password to access the application form.

Application form

The Data Safe Haven

UCL's Data Safe Haven (DSH) should be used for storing or processing sensitive data, including identifiable personal data and data that needs to be treated as special category for legal or contractual reasons.  The core Research Computing Platforms aren't suitable for use with this type of data.  If you think your research data may need to be in the DSH please contact infogov@ucl.ac.uk for advice. Please also see the Data Safe Haven User Guide & FAQs

The process for gaining access to the Data Safe Haven is described on the Data Safe Haven Assurance page.

Access to additional resources

Purchasing nodes on our clusters

Researchers can purchase nodes on our Myriad platform for their own use. This solution provides a dedicated computing resource without the need to queue or worry about setting up and maintaining hardware yourself. See the research computing documentation for more information about this option.

Due to the current restrictions on movement there may be significant delay in installing these nodes physically at our datacentre.  If your purpose is COVID-19 related we will attempt to be accommodating.

Other high performance computing services at UCL

There are several other HPC platforms run by other departments or institutes which you may be eligible to use; the following guide provides a quick overview of these services:

 Register of Research HPC Clusters at UCL

National research computing platforms

Researchers may be eligible to use national high performance computing platforms depending on their research domain and funding. See the HPC-UK website for details.