Science IT Computing
Science IT currently provides the following computing services to UZH researchers:
- ScienceCloud implements an Infrastructure as a Service (IaaS) approach to cloud computing. It can be used for the creation of ephemeral private computing systems and experimentation with new computing paradigms.
- ScienceCluster is a centrally managed multi-purpose cluster that provides access to multiple CPU and GPU resources. It also provides ScienceApps, a browser based portal to access ScienceCluster via a suite of interactive computational environments (Jupyter, RStudio).
- Supercomputer is a system for running large-scale high-performance parallel computations whose requirements exceed the ScienceCluster capacity.
Don't know which service is right for you?
The computing service selection guidelines below can help you decide which service best matches your research needs. Please note that these guidelines offer only very rough guidance. For further consultation on which service is most suitable, feel free to contact the Science IT Helpdesk.
How to get access
If your group already has a Science IT Computing service agreement in place:
- Contact your supervisor to learn who is the technical contact within your group
- The technial contact will then initiate your access
If your group is starting to use Science IT Computing services:
- To get onboarded contact the Science IT Helpdesk to schedule an initial meeting
- Topics discussed in this meeting include: Your use cases, services overview, terms and conditions (incl. costs), trainings, support provided, next steps
Terms and conditions
- Framework agreement covering all Science IT Computing Services
- Pay-per-use cost contribution model
- Usage within limits of regulations related to use of IT at UZH
- Access to Computing services restricted to UZH researchers
- More details (incl. costs & FAQs) are available in the UZH Intranet
- For further information contact the Science IT Helpdesk
Computing Service Selection Guidelines
Should I use the ScienceCloud or the ScienceCluster?
This decision tree may help you choose the appropriate solution for your needs. Please note that this tree is very general and offers only rough guidance.
What storage options are available?
- Cinder Block Storage
- Works exclusively with the ScienceCloud and works "out of the box" with a ScienceCloud tenant
- Must be attached to a ScienceCloud server instance to be used
- Cinder "volumes" act like virtual USB drives that "plug in" to virtual machine instances; volumes can only be attached to a single instance at a time
- Full access to files is allowed (upload, download, and editing)
- Swift Object Storage
- Offers accessibility from ScienceCloud virtual machines or from personal computers (both of which must be equipped with the appropriate software)
- The download/upload of files is possible, while editing files is not possible
- Swift Storage provides file space in what is termed "containers"
- Cluster Scalable Storage
- Works exclusively with the ScienceCluster
- Allows for an expansion of the default storage space provided with standard Cluster access
- If you need this option, contact us and inquire about "Scalable Storage" for the Cluster
- ZI Share + Backup
- Provides a remote file directory that's mirrored / backed-up to multiple locations
- The extra layer of security incurs an extra cost, as does restoring data from a back up
- ZI NAS
- "NAS" stands for "network attached storage"
- Works like a remote file directory that can be mounted to one (or more) personal computers or virtual machines at a time
- Requires additional configuration to work with ScienceCloud virtual machines
What are the options for archiving data?
Please see the descriptions of each service in the question titled "What storage options are available?".