Research Cloud Service

Description

The Reading Research Cloud is a platform for server hosting which runs virtual machine (VMs) on University owned servers, The platform is comprised of servers with a combination of: high end CPU, GPU, Memory and  SSD Storage. Also all the servers are connected by a reliable high speed network. This means that the platform is flexible enough to meet the users research requirements while being available at an affordable price.

The Research Cloud is managed by a self-service portal. Because of this the user can provision and configure servers to their own needs. We also offer a set of standard templates which can be used and adapted by the end user.

How to access

Initially, access to the Research Cloud can be requested by contacting your IT Business Partner. We are in the process of producing Self-service forms so that requests can be made via the IT Service Portal.

Research Cloud Pricing

Users can calculate the costs for using virtual machines or storage by combining the basic units below as required.

  • CPU – £68.00 per core per year
  • Memory – £34.00 per GB per year
  • Data Storage – £274.00 per TB per year
  • GPU – £125.00 per GPU per year (640 CUDA Cores)
  • Offsite backup for VM – £60.00 per TB per year

These costs have been certified for use with research councils. VAT will be charged for external entities.

The below table shows costs for some typical configurations which are comparable to costs on public clouds.

Typical VM Prices

Annual Cost VM Configuration
£136.00 1 core, 2 GB RAM
£272.00 2 core, 4 GB RAM
£544.00 4 core, 8 GB RAM
£1,088.00 8 core, 16 GB RAM
£2,175.00 16 core, 32 GB RAM
£4,351.00 32 core, 64 GB RAM

Research Data Storage Service

Description

Research Data Storage Service provides researchers with on campus access to 3 tiers of secure and resilient storage. Two of the three service options also offer off- site backup.

All storage is accessible from Windows, Linux and MacOS via SMB or NFS although we are unable to offer mixed access.

More information can be found on the IT Service Portal

How to access

Research Data Storage can be requested via IT Service Portal.

Costs

  Basic Silver* Gold
Snapshots No Yes Yes
Backups No Optional Add-on at £60 per TB per year. Yes (3 Months)
Suggested use General purpose research data which can be easily recreated or downloaded from another source. Research Data to be used with computing platforms or fast access via desktops. – Part of the Research Cloud. General purpose research data which is not easy to recreate, where additional recovery options are available.
Data protection level
One copy protected with parity drives. Two copies stored in one location on SSD computing tier with optional offsite cloud backup Two copies stored in two locations with offsite cloud backup
Cost per TB per year £103.00 £274.00 £421.00

*There is currently a 40TB limit on Silver storage to maintain performance.


Expired volumes

In order to ensure fair usage, lapsed storage which is not renewed will be marked as read-only for three months. After three months, the lapsed volume will be deleted along with all remaining data.

Research Data Archive

Description

The University of Reading Research Data Archive is a multidisciplinary online service for the registration, preservation and publication of research data produced or collected at the University of Reading.

University research staff and research students can register and deposit data in the Archive. Up to 20 gigabytes of data per project can deposited at no charge. Deposits greater than 20 GB may be made subject to a charge.

The Archive will enable you to comply with the University Research Data Management Policy and any relevant funder policy for the long-term preservation/sharing of research data. In particular you can:

  • Create and publish an online metadata record for a digital and/or non-digital dataset;
  • Generate a unique permanent Digital Object Identifier (DOI) for your dataset, so that it can be easily cited and linked to;
  • Deposit digital data files and related documentation for long-term preservation and access;
  • License your data and use access settings to control how your data files can be accessed and used.

How to access

The Research Data Archive is managed by Research Services and can be accessed via the RES website.

Note: This service is not offered directly by IT.

Reading Academic Computing Cluster Service

Description

The Reading Academic Computing Cluster (RACC) is a Linux cluster which provides resources for interactive research computing and batch jobs submissions to the University of Reading staff and students.

The cluster consists of login nodes and a larger number of compute nodes. On a typical cluster, login nodes are used only for preparing and submitting batch jobs. Here, we provide several load-balanced login nodes, equipped with multi-core processors, large memory and fast network connection to the research data storage volumes, such that they can also be used for interactive research computing, such as data analysis and visualization, or code development and testing.

In batch mode the tasks are described in a script and submitted to the scheduler, to be run on one of the compute nodes without further user intervention.  The compute nodes provide over 500 CPU cores for serial and parallel batch jobs. The default time limit for a batch job is 24 hours, but when it is justified, running longer batch jobs is also possible. When justified by the project requirements, it is also possible to use compute node allocation for interactive processing.

In the future, approximately 300TB of local scratch storage will be provided.

A comprehensive list of installed software can be found here.

How to access

The RACC can be accessed by connecting to cluster.act.rdg.ac.uk:
ssh cluster.act.rdg.ac.uk

Costs

The majority of cluster resources are available free of charge, subject to fair use policy. Details of this policy will follow. Dedicated paid-for resources can be purchased and integrated with the cluster.

 

Off Site Cloud Storage

Off Site Cloud Storage

OneDrive for Business

Description

OneDrive for Business is a managed cloud storage that allows users to store and share files and folders online.

How to access

Simply login to Office 365 to gain access to OneDrive for Business:

loginbutt

 

USEFUL LINKS/GUIDES

Microsoft has provided some user guides to help you get up and running with OneDrive for Business:

USE OF ONEDRIVE

Ensure there are no contractual restrictions on use of Cloud Storage for the work you wish to store in One Drive. Use OneDrive for collaboration but be mindful that any documents used by a team will need to be moved should the owner of the document leave. Please read the University of Reading’s usage requirements regarding OneDrive.

 

  OneDrive For Business
Cost Free of charge for all Staff & Students
Default capacity 5TB
Integration with UoR systems Yes – uses UoR login
Approved for storage of UoR  data Yes
Is data stored in the EU Yes
Maximum file size (for individual file) 15GB
Supported Operating systems Windows (App & Web)

MacOS (App & Web)

Linux (Web only)

Ability to access files via App, without having to Sync entire content Windows 10 (from Q2 2018) tbc

MacOS (10.9 and above)

Data can be shared with staff/students at UoR? Yes
Data can be shared with users outside of UoR Yes
Data can be shared with users who don’t have an account with the provider Yes

 

 

Dropbox for Business

Description

Dropbox for Business is a managed cloud storage that allows users to store and share files and folders online.

How to access

Go directly on dropbox.com and subscribe using a P-card

 

Use Of Dropbox for Business

Ensure there are no contractual restrictions on use of Cloud Storage for the work you wish to store in Dropbox. Use Dropbox for collaboration but be mindful that any documents used by a team will need to be moved should the owner of the document leave. Dropbox is only recommended where OneDrive for Business does not meet requirements.

 

  Dropbox Teams
Cost £15 per month (up to 3 users)
Default capacity As much as you can use
Integration with UoR systems No – requires separate Dropbox account
Approved for storage of UoR  data Should not be used for high risk personal data or sensitive information
Is data stored in the EU No
Maximum file size (for individual file) No limit if uploaded via Dropbox App.

20GB if uploaded via web browser

Supported Operating systems Windows (App & Web)

MacOS (App & Web)

Linux (Web only) tbc

Ability to access files via App, without having to Sync entire content Windows Vista and above

MacOS (10.9 and above)

Data can be shared with staff/students at UoR? Yes
Data can be shared with users outside of UoR Yes
Data can be shared with users who don’t have an account with the provider Yes

Managed Desktop Service

Note: the following desktops are remote desktops and users need to connect with their PC / Mac. ACT supports the services, but do not provide the personal desktop hardware.

Linux Desktops

Description

The Linux Desktop service uses NoMachine NX to provide access to Linux desktop environment software running on a set of servers, via client software running on office desktop computers.
The main purpose of the Linux Desktop Service is to facilitate the use of the Research Processing service for Windows and Mac users.
Using the NoMachine software, you’ll have to first connect to nx.reading.ac.uk. Read more…

The only Linux desktop applications supported by IT are the following terminal emulators: gnome-terminal and xterm.  Other Linux desktop applications may be installed but no support for their use is available from IT.

How to access

The Linux desktop service is available to all staff. If you have any issues, please contact the IT Service Desk.

See user’s guide.

Costs

The Linux desktop service is free of charge.

Windows Desktops

Description

The Windows Desktop service uses Microsoft Remote Desktop to provide access to a managed Windows desktop environment running on a set of servers, via client software running on office desktop computers.
The main purpose of the Windows Desktop Service is to enable access to Windows applications from a Linux environment.

How to access

The Windows desktop service is available to all staff. If you have any issues, please contact the IT Service Desk.

Read the user’s guide to access the Windows Desktop Service.

Costs

The windows desktop service is free of charge.

ACT software list

ACT software list

Overview of Installed Software

This page contains information about the software that is available on the academic computing cluster, the met-cluster and the NX nodes. The lists are not exhaustive as we restricted them to software that isn’t available by default in Centos 7 Linux and focus here on scientific software and programming tools.

 

Reading Academic Computing Cluster (RACC)

On the RACC software is both installed in default system paths from software package managers (rpms) and as environmental modules, which have to be loaded first.

Software installed from rpms

The default versions of these software packages are already in the path and the ‘module load’ command is not needed.

Package Description Commands
R Language for statistical computing and graphics,
https://www.r-project.org
R
rstudio GUI environment for R,
https://www.rstudio.com/
rstudio
octave Numerical programming suite compatible with Matlab,
https://www.gnu.org/software/octave/
octave
gcc GNU Compiler Collection,
https://www.gnu.org/software/gcc/
gcc, g++, gfortran
git Version control system,
https://git-scm.com
git
GSL GNU Scientific Library,
https://www.gnu.org/software/gsl/
boost Peer-reviewed portable C++ source libraries,
https://www.boost.org
lapack Linear Algebra Package,
http://www.netlib.org/lapack/
blas Basic Linear Algebra Subprograms,
http://www.netlib.org/blas/
gdal Geospatial Data Abstraction Library,
https://www.gdal.org
zlib A Massively Spiffy Yet Delicately Unobtrusive Compression Library,
https://zlib.net
grib-api ECMWF application program interface accessible for encoding and decoding data in grib format,
https://confluence.ecmwf.int/display/GRIB/Home
e.g. grib_ls, grib_info, grib_get
curl Command-line tool and library for transferring data,
https://curl.haxx.se
curl
fftw C subroutine library for computing the discrete Fourier transform (DFT),
http://www.fftw.org
udunits C library for units of physical quantities and a unit-definition and value-conversion utility,
https://www.unidata.ucar.edu/software/udunits/
udunits
expat A stream-oriented parser,
https://libexpat.github.io
proj A generic coordinate transformation software,
https://proj4.org
proj
ncl NCAR Command Language,
https://www.ncl.ucar.edu
ncl
nco netCDF Operators’
http://nco.sourceforge.net
e.g. ncks, ncecat, ncwa
ncview netCDF visual browser,
http://meteora.ucsd.edu/~pierce/ncview_home_page.html
ncview
cdo Climate Data Operators,
https://code.mpimet.mpg.de/projects/cdo
cdo
mplayer Movie Player,
http://www.mplayerhq.hu/design7/news.html
mplayer
cmake Tools to build, test and package software,
https://cmake.org
cmake
rsync File transfer utility,
https://rsync.samba.org
rsync
rclone A command line program to sync files and directories to and from cloud storage including OneDrive,
https://rclone.org
 rclone
firefox Web browser,
https://www.mozilla.org/en-GB/firefox/
firefox

 

Software available using environmental modules

If you’re not familiar with the ‘module’ commands, please see Accessing Software on the Cluster for information about using environmental modules.

Package Description Commands
ENVI (licensed for v. 5.5 and onward) Image analysis software
https://www.harrisgeospatial.com/Software-Technology/ENVI
envi
IDL Scientific programming language,
https://www.harrisgeospatial.com/Software-Technology/IDL
idl
Matlab Numerical computing software and programming language, see also: Running Matlab Scripts as Batch Jobs
https://uk.mathworks.com/products/matlab.html
matlab
Canopy python (2.7 and 3.5) Enthought python distribution and analysis environment for scientific and analytic computing, see also: Python on the Cluster
https://www.enthought.com/product/canopy/
python
Anaconda Python (2.7 and 3.5) Anaconda python distribution and analysis environment for scientific and analytic computing, see also: Python on the Cluster
https://www.anaconda.com
python
cuda NVIDIA parallel computing platform and programming model,
https://developer.nvidia.com/cuda-zone
nvcc
gcc GNU Compiler Collection,
https://www.gnu.org/software/gcc/
gcc, g++, gfortran
intel Intel Parallel Studio XE for Fortran,
https://software.intel.com/en-us/parallel-studio-xe
ifort
pgi PGI Fortran compiler and tools,
https://www.pgroup.com
e.g. pgf90, pgcc, pgunzip
solstudio Oracle Developer Studio compilers and tools,
https://www.oracle.com/technetwork/server-storage/developerstudio/overview/index.html
e.g. cc, CC, f90, f95
mpich High performance implementation of the Message Passing Interface (MPI) standard,
https://www.mpich.org
e.g. mpicc, mpifort, mpiexec
openmpi Open source implementation of the Message Passing Interface (MPI) standard,
https://www.open-mpi.org
e.g. mpicc, mpifort, mpiexec
nag NAG library, a collection of robust, documented, tested and maintained numerical algorithms,
https://www.nag.co.uk/content/nag-library 
NAG library on RACC
netCDF Libraries and tools that support the creation, access and sharing of array-oriented scientific data,
https://www.unidata.ucar.edu/software/netcdf/
e.g ncdump, ncgen, nccopy
hdf5 Technology suite for the management of extremely large and complex data collections,
https://support.hdfgroup.org/HDF5/
e.g h5dump, h5diff, h5ls

 

Met-Cluster

The met-cluster is now a legacy service. While it is still available, software on it can be accessed via the ‘module‘ command. Some of the software and packages that are listed for the RACC are also available for the met cluster, although these are older versions and are no longer maintained. For up to date software and libraries, please switch to the RACC as there are no more updates or additions planned to the met-cluster software.

 

NX nodes

The NX nodes use the NoMachine NX Enterprise software to provide a virtual linux desktop. The software is only installed in default system paths from software package managers (apt).

Package Description Commands
git Version control system,
https://git-scm.com
git
curl Command-line tool and library for transferring data,
https://curl.haxx.se
curl
ncview netCDF visual browser,
http://meteora.ucsd.edu/~pierce/ncview_home_page.html
ncview
rclone A command line program to sync files and directories to and from cloud storage including OneDrive, see also: Transferring files to OneDrive for Business via rclone
https://rclone.org
 rclone
rsync File transfer utility,
https://rsync.samba.org
rsync
firefox Web browser,
https://www.mozilla.org/en-GB/firefox/
firefox

Packages installed for Meteorology

Ubuntu  repositories:

cmake,ghostscript,gimp,git,gv,imagemagick,latexdiff,latex2rtf,libreoffice,lynx,pdftk,php,r-base,rcs,subversion,texlive-full,texmaker,w3m,vlc,xfig,xournal

Other repositories:

chrome

Packages installed for Computer Science

Ubuntu  repositories:

cmake-qt-gui,eclipse,git-gui,gedit,meld,nano,vim,gcc,clang,dnsutils,strace,doxygen

Other repositories:

atom,eclipse,visual_studio_code

Direct install from archive file

archi

WordPress websites for research groups and projects

WordPress websites for research groups and projects

ACT is providing a WordPress environment to host your research websites.

WordPress is a web Content Management System (CMS) that allows you to create, edit your website pages as easily as editing a document in a text editor.

This service is free of charge.

If you want to create a website, contact the Research Communication team (Caroline Knowles and Jo Stone). Read the full procedure description below.

The default website URL will be: https://research.reading.ac.uk/yourwebsite. However, you can have a domain name that will redirect to this URL.

Here are a few examples of websites hosted on this server:

Read more: starting with wordpress.

WordPress Research site: Request and Approval Process

  1. Researcher contacts RCE detailing who they are, why they need a website (the request may come via IT or Content team, who should forward it on to RCE).
  2. RCE logs the request on website development spreadsheet.
  3. RCE responds to researcher (wherever possible, within five days) to discuss the request. This may take place via email, phone call or preferably by meeting to discuss the requirements and what kind of web presence would be most suitable).
  4. RCE approves a WordPress site or advises on an alternative option.
  5. If approved, RCE sends a request to IT to set up the site, URL and a list of required users.
  6. RCE arranges WordPress training with researcher (to be provided by trainer in Content team, currently being recruited).