user you can request a virtual machine hosted on our private cloud to carry out analysis in your own Linux environment, just like Amazon Web Services or Microsoft Azure.
You can use your custom VM to do many things, such as running a web service to serve data and analysis to your community, to submit bigger jobs to our CyVerse UK
HPC powered by Condor, or as a virtual laboratory or development ecosystem to share with your collaborators using a platform like the Genomics Virtual Laboratory (GVL). There are many possibilities!
What resources can I get?
We have a range of VM options available that you can request:
- small blank ubuntu xenial: 16GB RAM and 2vCPU
- medium blank ubuntu xenial: 32GB RAM and 4vCPU
- large blank ubuntu xenial: 64GB RAM and 8vCPU
- small blank centOS7: 16GB RAM and 2vCPU
- medium blank centOS7: 32GB RAM and 4vCPU
- large blank centOS: 64GB RAM and 8vCPU
- small ubuntu xenial: 16GB RAM and 2vCPU + R/python/nginx installed
- medium blank ubuntu xenial: 32GB RAM and 4vCPU + R/python/nginx installed
- large blank ubuntu xenial: 64GB RAM and 8vCPU + R/python/nginx installed
- small blank centOS7: 16GB RAM and 2vCPU + R/python/nginx installed
- medium blank centOS7: 32GB RAM and 4vCPU + R/python/nginx installed
- large blank centOS: 64GB RAM and 8vCPU + R/python/nginx installed
You will be given sudo rights on the machine and will therefore be able to install additional software and packages you may require. You will not be able to modify the networking rules on the VM, and only the following ports will be made available by default: 22 (SSH), or 22(SSH), 80 (HTTP) and 443 (HTTPS). Any other ports you may require will need to be approved and set up ahead of time, as we would need to know the services you wish to run on those ports, and the services would need to be security tested.
Additionally, you can request for more software to be installed at the time of deployment.
What storage space do I get?
Currently, you can request one of the VM recipes above with a 100GB volume.
Our policy for free machines is first-come first-served and this comes with restrictions on CPU and storage availability. We reserve to adjust our policy at any time, considering the interest and the resources users ask for, to manage contention on the system.
If you need more space, we request that you get in touch and give us some more information about your project requirements and additional resources. We can then form a better picture of how our resources can be managed, and any cost recovery can be put in place. We love to work collaboratively, so if you need HPC resources for a grant, then please consider including our National Capability in your costings by contacting us to work out requirements.
What’s the process to request a VM?
We request you contact us
to discuss your project and the resources required for your research process. After you contact us we’ll get back to you and:
- if you requested one of the standard size/software above, we’ll spin the machine up immediately
- if you would like different OS/additional software/different computational resources, we’ll likely ask you for more details
In both cases, we’ll send you a Service Level Agreement
to sign detailing the resources we agreed on. Once the machine is up and running we’ll email you how to log into it with a public SSH key. We’ll also send you some documentation on how to do so if you need it.
For security purposes you should create the key(s) yourself on the system(s) that you will be using to log in to your VM and share ONLY the public key with us. Also don’t share the password we’ll send you with anyone!
Once your key is created, please do not share your private key with anyone. You may add collaborators with different key pairs, but you’ll have to contact us so we can register them on the system.
We reserve the right to access to your VM for system administration purposes, but we guarantee that we won’t use it unless you ask for support.
Once you are logged in and working, the VM is yours!
How can I manage my data?
You can ask for up to 100G volume space on a per user basis to be attached to your virtual machine. We suggest you treat this storage as a scratch directory: we have mechanisms in place so that you won’t lose data, but volumes are not often backed up regularly and we can’t assure 100% they won’t be compromised in case of technical issues. You are responsible for the data that is on your system. Any data infringing copyright or other legalities are completely under your responsibility. EI cannot be held liable for any infringements occurring on our private cloud systems. We will fully comply with authorities in any legal cases arising from the use of our VMs.
We can also provide you with a bucket in our development object storage if you request it, mind it’s only for development at the time being!
We recommend that you set up a CyVerse
account before requesting a VM. This means you will immediately be able to use iRODS to transfer data with icommands to and from the CyVerse UK
Data Stores, adding a layer of collaboration and security to your data. Once you finish running your analyses you can push your results to the Data Store, and you can then easily share your data more widely with collaborators, other researchers, and the public.
How can I gain access to public data?
Open science comes with the ability to download and use potentially huge public datasets within your own research, improving the quality, transparency, and recognition of data. We have a range of open datasets available for our key communities, and these can be mounted on your VM for you to use right away, without needing to download them ahead of time. These datasets are currently being prepared, so if you have any datasets you know you will need, then let us know and we can add them to our Data Commons.