In Autumn 2020 we will be upgrading the operating system on BlueBEAR and BEAR Cloud and CaStLeS Virtual Machines. Some applications will no longer be supported. Please BEAR Applications Versions for details.
Cloud means many things to many different people. BEAR Cloud is an onsite private cloud operated as part of the wider BEAR service offering.
At the simplest level, BEAR Cloud is a bunch of powerful computing resources (dedicated to supporting computationally or data intensive research) that can be deployed flexibly and changed easily to meet evolving user needs. The entire infrastructure is housed on campus; a characteristic which is particularly important for those dealing with large datasets or sensitive data. It is also highly efficient and cost effective in comparison with current commercial cloud offerings.
Built with a major investment by the University (including part of a multi-million pound investment for Life Sciences) and using some innovative technology, Advanced Research Computing operates the underpinning infrastructure and base services. BEAR Cloud has been developed to support different types of workload to those operated on the BlueBEAR supercomputing facilities. Resouce is available both as part of general and CaStLeS facilities to support research workloads from across the University. Unlike many public cloud providers, the services we provide on BEAR Cloud don't over allocate CPU or memory, so when you get a CPU core, its dedicated to your workload.
Our principle offering on BEAR Cloud is to provide high powered virtual machines These servers can have different operating systems – different varieties of Linux for instance – and can be tailored for applications that have been developed for particular environments. The servers we offer are specifically to support demanding research workloads, this could be computationally or IO driven workloads and as far as possible are connected to our research data storage systems.
As we develop BEAR Cloud, we're looking to support more containerised and user-managed workloads on the service.
Who can use the service?
BEAR Cloud is available to researchers across the University however we undertake a technical review of your requirements before providing access to ensure BEAR Cloud is the correct offering for you. All use of BEAR Cloud must be attached to a BEAR project. Where users are requesting access to CaStLeS resources, this will additionally be reviewed using the standard CaStLeS request process.
What types of virual machine can I get?
Our VM sizing is coded by colour, information on the standard VMs we have is available.
How does BEAR Cloud work?
BEAR Cloud uses the same hardware platform we have for other compute services but it not part of the BlueBEAR supercomputing facility. Instead the BEAR Cloud systems use OpenStack, an open source private cloud set of software to run the service. You can find out more information about the technology used in BEAR Cloud.