Section 2.3 The ABoVE Science Cloud (ASC)
Referenced on page A.4-8 in NASA Research Announcement for Terrestrial Ecology: Airborne Campaign For ABoVE NNH16ZDA001N-TE
The NASA Center for Climate Simulation (NCCS) has partnered with the NASA Carbon Cycle and Ecosystems Office (CCE Office) to create a high performance science cloud for this field campaign. The ABoVE Science Cloud combines high performance computing with emerging technologies and data management with tools for analyzing and processing geographic information to create an environment specifically designed for large-scale modeling, analysis of remote sensing data, copious disk storage for “big data” with integrated data management, and integration of core variables from in-situ networks. The ABoVE Science Cloud is a collaboration that promises to accelerate the pace of new Arctic science for researchers participating in the field campaign. Furthermore, by using the ABoVE Science Cloud as a shared and centralized resource, researchers reduce costs for their proposed work, making proposed research more competitive. The ABoVE Science Cloud is available to all ABoVE projects. To join the ABoVE, see here >>.
The ABoVE Science Cloud:
- Provides a shared set of computational and data resources to the ABoVE science team.
- Enables access to large, common data sets (both observation and model) that are relevant to the ABoVE research goals.
- Provides a system and collaboration environment by which results may be quickly and readily shared to the ABoVE research community and ultimately to decision makers.
- Enables researchers to propose larger problems, analytics, and more science than they could address using the capabilities of typical computer workstations.
- Provides tailored computational, analysis, and data management environments to meet the needs of the individual science projects.
- Supports researchers with comprehensive services to facilitate the use of advanced information technology, creation of metadata and documentation, and archival of finalized products.
ABoVE Science Cloud Capabilities:
The architecture of the ABoVE Science Cloud combines high-performance computing and cloud technologies to create an extremely flexible and agile platform with the following capabilities:
- Data services nodes are used for persistent services that need to be accessible from outside the science cloud. Virtual systems (either virtual machines or containers) will be deployed on the data services nodes to create web services that can be easily backed up and migrated for high availability.
- Database nodes in the science cloud can be used for high performance, highly available database requirements.
- Remote visualization nodes enable server side graphical processing and rendering of data.
- High performance compute nodes will be used for itinerant or elastic processing requirements capable of supporting 100’s of virtual machines. In the event that a specific researcher or even a web service needs additional processing capability for a short time period, the cloud can be configured to provision additional virtual machines to run processing in parallel and greatly reduce the time to solution.
- High-speed and high-capacity storage are configured for use throughout the science cloud for a variety of purposes, including redundant, snapshotted home directories for code development, scratch space for scientific processing, and large data storage repositories for public data. Three pre-ABoVE funded investigations are creating data products using MODIS and Landsat data sets currently stored in the science cloud. Proposing investigators can expect to request support and assistance to ingest and manage additional data sets of high importance for ABoVE research (e.g. commercial high-resolution satellite imagery; core observational data identified in the Concise Experiment Plan).
- Data products on the ASC – Data products can be added to the ASC as needed based on the requirements of ABoVE investigators and the needs of the overall ABoVE Science Team. A current list of available products in the ASC can be found here (note that the DigitalGlobe archive is not included in the associated table).
Science Cloud Setup Instructions >>
A monthly webinar is held on the 3rd Friday of the month at 1pm Eastern time.
Topics of interest to ASC users are discussed including how to use the system and examples from users. Presentations are recorded and made available. Go to archive >>
Contact Liz Hoy to submit a topic or to find out more information.