This page details different types of research data environments available to those at the College. Use the tabs below to get general details on each data environment and then view the PDF for more information on each enclave. 

Summary of Key Research Data Environments

Secure Enclaves

Secure Enclaves

An environment for groups dealing with identifiable and de-identified sensitive data. It is designed to give stakeholders assurance that groups meet the Information Governance requirements that they are subject to (e.g. GDPR).

Owned by: Faculty of Medicine
Hosted location: On-premise (ICT Data Centre)
Key contact: secure-enclaves@imperial.ac.uk

Facilities offered:

  • Each group is given a secured area of the College network (an enclave) that is isolated from all other areas of the network (including other groups in the environment).
  • Access to servers is through a controlled and audited management service.
  • Users, workstations and servers are subject to a common Information Governance Policy.
  • Enclaves can contain servers running Linux or Windows operating systems. Servers may be physical, virtual or cloud servers, providing they meet the IG policy requirements.
  • BitLocker 256 encryption is used for data at rest on the server disks. Symantec desktop encryption is also used when sending data over the network.
  • The ICT backup service is used for all data stored by research groups.

View the Secure Enclaves Sharepoint site for further information. You will need to be logged into your College account to access this.

BDAU SE

Big Data & Analytical Unit Secure Environment (BDAU SE)

An ISO 27001 certified research environment, compliant with NHS Data Security and Protection Toolkit, located at the College data centre.
It offers multi-user batch processing and an interactive GUI with database options, GPU/TPU facilities, and open source or proprietary software in a fully audited environment.
The environment is suitable for ongoing research projects involving the use of de-identified personal data which needs to be handled in a secure way to ensure compliance with regulations such as GDPR, or the data security requirements as set by data providers.

Owned by: Institute of Global Health Innovation - BDAU
Hosted location: On-premise (ICT Data Centre)
Key contact: BDAU Team: bdau@imperial.ac.uk

Facilities offered:

  • A standard operating/access model
  • Secure, fault-tolerant data storage encrypted to AES256
  • Open-source or proprietary analysis software
  • Multi-user batch processing
  • GPU / TPU facilities
  • Network speed: 10 Gbps
  • Ubuntu LTS based environment
  • Multi-layered multi-factor authentication

Data Science Institute (DSI)

Data Science Institute (DSI) Cloud Environment

Provides modern and future-proof cloud computing services for users and selected collaborators. Hosting of applications for processing sensitive data.
Owned by: Data Science Institute
Hosted location: On-premise (DSI private cloud environment)
Key contact: Dr Florian Guitton (f.guitton@imperial.ac.uk) Mr Oliver Nixon (o.nixon@imperial.ac.uk)

Facilities offered:

  • Dedicated hardware for servers and network
  • Segregated physical network
  • Private virtual networks
  • Private virtual machines - user cloud environment
  • Fault-tolerant storage
  • Secure web applications
  • MongoDB database cluster
  • GPU cluster
  • TPU training and inference equipment
  • Open-source software and hardware solutions
  • Ubuntu LTS-based installation
  • Full hardware encryption
  • Network speed: 100 Gbps backbone, 20 Gbps minimum to the node

ICT Research Computing Services

A multi-user batch processing environment, supporting a diverse set of high-throughput and high-end workloads. RCS provides bulk compute processing and high-volume data storage and processing capability to the College's research community.

Owned by: Central Faculty - ICT
Hosted location: On-premise (ICT Data Centre)
Key contact: rcs-support@imperial.ac.uk

Facilities offered:

  • 80k compute cores in total, with the current maximum configurable job size of 2560 core with 40TB RAM
  • 160xRTX6000 GPU
  • 153 GPU of older specifications
  • Storage throughput of 25GB/s
  • Research Software Engineering support/consultancy/training available
  • Research Data Store service
  • Site-to-site data transfers with Globus web service