Sri Harsha Vathsavayi
The amount of data collected by scientific data repositories is growing day by day. Data repositories are growing rapidly. With the rapid increase in the growth of data sets the computational capacity needed for processing and analyzing the data sets is also increasing. It is no longer feasible for the users to download the data to local machines and process and analyze the data. To overcome this problem, scientific communities are moving towards building virtual research environments (VREs) on top of the data repositories. The VREs are seen as a science gateways to the research communities for accessing data, creating new data products, sharing relevant findings with colleagues and publishing findings. Moreover, they are an important platform for replicating research experiments and workflows. This abstract aims at demonstrating the VRE being developed for the marine research community.
The SeaDataNet project (https://www.seadatanet.org/) is a primary entrypoint for marine researchers to find and access data. They provide a pan european infrastructure that collects data from more than 100 data centers. To provide a common scientific gateway for the marine research community the project has partnered with EUDAT CDI (https://eudat.eu/) in the SeaDataCloud project to build a VRE that can be seamlessly accessed by the researchers across Europe. The main requirements for the SeaDataCloud VRE are: to provide on demand computing and data access to researchers across Europe, make it easy for the users to bring their own data to the environment and subsequently download the results, integrate various existing monolithic services to the environment and the users should be able to use their own existing organization credentials to access the system. This abstract presents the high-level architecture and technical choices selected for fulfilling the SeaDataNet VRE requirements to serve users across Europe.
The VRE infrastructure will be built using OpenStack cloud and container management platforms like RANCHER to handle high demand from users across Europe. The service integrates the user’s personal cloud storage B2DROP (https://b2drop.eudat.eu/) for securely bringing his/her own data sets to the VRE service and at the same time seamlessly share the results between VRE service and the user’s personal computer. The VRE is based on micro-service architecture style and the existing monolithic services are converted into micro-services and integrated to the VRE service. To make it easy for the researchers across Europe to access the VRE service, the service integrates with B2ACCESS AAI (https://b2access.eudat.eu/), which can be accessed using different organization accounts across Europe and also using user’s orcid account.
Summarising, the technical solutions used in building the SeaDataCloud VRE are applicable to other research communities as well.
We present the virtual research environment being developed for the marine research community. The virtual research environment is developed as a collaboration between SeaDataNet community and EUDAT CDI. The virtual research environment will offer a common scientific gateway for the marine research community. It will provide a workplace where scientific users will be able to select, filter, aggregate, synthesize and analyse large volume of reference marine observations. The architecture and technical solutions chosen for realising the virtual research environment are presented. These solutions and architecture could be used as reference solutions for building virtual research environments for new research communities.
|Type of abstract
Sri Harsha Vathsavayi