Eddie

A high-performance Linux computing cluster to run analysis faster than a desktop computer.

Eddie is the University's Research Compute Cluster. It runs a Linux operating system and can analyse large amounts of data faster than a desktop computer.

Eddie can help you work with very large datasets of any type. You can use it for simulation and modelling, as well as for data analysis, manipulation, mining and visualisation.

Eddie

Features:

  • Work can be run in parallel taking advantage of multiple compute cores
  • Priority and exclusive access to resources is available for funded projects
  • A number of applications are installed on Eddie and you can install your own if required
  • Eddie connects with DataStore, the University's file store for research data, making it easy to transfer your data for analysis
  • You can learn how to use Eddie even if you have little or no experience of cluster computing, as training, support and consultancy services are available

The Institute of Genetics and Molecular Medicine and the Roslin Institute have their own sections of Eddie. Access to these reserved areas for users from these institutes is managed locally.

Find out more about Eddie Getting Started On Eddie

Free or paid for

Standard allocation: Free

Paid-for options:

  • Priority - ensures your jobs go to the head of the queue for processing, Cost per core hour of £0.02, which is £176 per core year
  • Ringfence – gives you exclusive access to one or more of the computing nodes for a set time, Cost per core hour of £0.02, which is £176 per core year
  • Storage – get additional storage space, £400 per terabyte per year
Find out more about Eddie costs

Product Features

  • High performance computing cluster
  • Over 7,000 cores
  • No cap on computing processing power
  • Option to purchase exclusive use of a computing node for a set time
  • Option to purchase additional space
  • Default 2TB temporary (28 days) storage shared by all users
  • Capacity to use your own software applications
  • Online guidance
  • Weekly drop-in clinics
  • Training course
  • Email and phone support
  • Intel processor based cluster running 64-bit Linux 
  • Each user can run thousands of jobs and subtasks at once
  • Option to purchase higher priority for processing job
  • Default 200GB storage space for research groups (on application)
  • Default 1GB storage for each individual user
  • Standard applications available: Anaconda, Java, Matlab, Python, R, TensorFlow
  • Data stored in the UK on University of Edinburgh servers
  • Data on Eddie is backed up nightly, but we recommend that DataStore is used to store ‘golden copy’ data as it is resilient to the loss of a complete data centre whereas the Eddie Filesystem is not.

Applicable Disciplines

  • Agricultural and Biological Sciences
  • Business, Management and Accounting
  • Computer Sciences
  • Energy
  • Health Professions
  • Material Science
  • Neuroscience
  • Physics and Astronomy
  • Arts and Humanities
  • Chemical Engineering
  • Earth and Planetary Sciences
  • Engineering
  • Immunology and Microbiology
  • Mathematics
  • Nursing
  • Psychology
  • Biochemistry, Genetics and Molecular Biology
  • Chemistry
  • Economics, Econometrics and Finance
  • Environmental Science
  • Medicine
  • Pharmacology, Toxicology and Pharmaceutics
  • Social Sciences
  • Veterinary

Case Studies

Related Training

Terms & Conditions

Specific:

General:

  • The Computing Regulations govern the use of all University computing and network facilities by staff, students and anyone else authorised to use them

Technical Requirements

Skills

  • Required:
    • Basic Linux experience
    • Mandatory training is required by some Schools or units: check with your School research officer 
  • Recommended:
    • Experience of using Linux clusters
    • Attending Introduction to Eddie course
    • Linux courses (in the University run by IS)

Support and Training Available