The Elastic Cloud Computing Cluster (EC3) is a platform that allows creating elastic virtual clusters on top of Infrastructure as a Service (IaaS) providers, either public (such as Amazon Web Services, Google Cloud or Microsoft Azure) or on-premises (such as OpenNebula and OpenStack). Through a 'job wizard' interface, the user can configure the virtual cluster with a predefined set of applications that will be deployed in the clouds underpinning the EGI Applications On Demand infrastructure. The installation and the configuration of the cluster are performed by means of the execution of Ansible receipts. The cluster configured by EC3 is private: as soon as it is configured the user will have root access to the environment, and can setup and configure the cluster installing additional libraries and software to their needs.
- Federated authentication through the EGI Applications on Demand service.
- Web based access.
- Interoperability with most widely used IaaS cloud technologies
- Wizard interface to streamline the configuration and the deployment of the virtual cluster on top of Infrastructure as a Service (IaaS) providers.
- Several Ansible receipts are available to deploy applications and tools in the cluster nodes.
- Nodes of the clusters will be self-managed by CLUES. Working nodes will be undeployed when they are idle.
- Use of ‘per-user subproxies’ from EGI to access X.509-protected resources on the users’ behalf.
- Access to the HNSciCloud commercial cloud service providers for limited scale usage, via a voucher of €250,00 euros.
- User support is available by an international network of consultants.
Scientific applications already available for access:
- Life Science (NAMD)
Tools available for access:
- OSCAR (Open Source Serverless Computing for Data-Processing Applications)
- A framework for efficiently supporting on-premises FaaS (Functions as a Service) general-purpose file processing computing applications.
- ECAS (ENES Climate Analytics Service):
- A complete environment enabling scientific end-users to perform data analysis experiments on large volumes of multidimensional data, by exploiting a PID-enabled, server-side, and parallel approach. It relies on Ophidia, a HPDA framework for eScience used to perform scientific data analytics by means of HPC paradigms and in-memory based big data approaches, and on JupyterHub, to give users access to ready-to-use computational environments and resources. Thanks to the Elastic Cloud Computing Cluster (EC3) platform, researchers will be able to rely on the EGI Cloud Compute service to scale up to larger datasets taking advantage of the underlying Infrastructure.
- SLURM (SLURMaaS):
- A service to deploy self-managed and customised SLURM clusters as a service with additional capabilities to support specific hardware backends in the EGI Federated Cloud. SLURM clusters are self-managed so they shrink or grow automatically (up to a predefined maximum quota) according to the workload.
- EKaaS (Elastic Kubernetes as a Service)
- A service to deploy self-managed and customised Kubernetes clusters as a service with additional capabilities to support specific hardware backends in the EGI Federated Cloud. EKaaS facilitates users to deploy not only the processing back-end in the form of a Kubernetes cluster but a set of applications as Helm charts. EKaaS clusters are self-managed so they shrink or grow automatically (up to a predefined maximum quota) according to the workload.
To submit an order request for this service, please register to the EOSC Portal Marketplace with the EGI AAI Check-In service.
Vouchers allow researchers to test different configurations or different services to choose the best offer for their needs.
Use the default quota granted by EGI to any authorised users
|Number of CPU Cores||4|
|Total amount of RAM||8|
|Number of CPU Cores: Up to||8|
|Total amount of RAM: Up to||32|
|Storage capacity: Up to||200|
|Number of days||365|
Places and languages
Related Infrastructures and Platforms
- EGI Applications on Demand
- EOSC-hub project
- Research Affiliation Needed