Princeton Climate Analytics (PCA) data products optimally merge a wide range of gauge, satellite, and reanalysis datasets, with advanced climate and hydrological modeling allowing to provide the most comprehensive and accurate past, present, and future climate data for risk analysis, business intelligence, and more, through a user-friendly and intuitive platform. For this purpose, the PCA platform aggregates highly precise climate data and prepares it for accurate forecasting and risk analysis.
These data and scripts were initially deployed to Princeton University’s supercomputer, which provided an excellent environment for research but offered no possibility for third-party access and full-scale commercial use. To enable a market launch, PCA needed to migrate its platform to the cloud.
The massive amount of data due to the long time series and global scope, as well as the complexity of the analytics scripts, were the main challenges of the migration.
The ELEKS team set up the cloud environment and migrated the monitor applications and historical data from Princeton University’s supercomputer to Amazon Cloud (AWS). Also, our data science experts checked the analytics models, deployed them to the new environment and made sure they ran smoothly after the AWS migration. Altogether, we have migrated two climate monitors, 21 analytics scripts and nearly 70 years of global precipitation data.
After the AWS migration, PCA data has been successfully applied on a market scale for modeling soil moisture and evaporation, estimating water resources and driving dynamic ecohydrological models for regions across the world.
