Igor Karuza
With 15+ years of experience in application development, database design, and information system architecture, I currently serve as a Product Manager, Architect, and Development Lead at IGEA, specializing in complex geoinformation systems. For over four years, I have been dedicated to advancing geoinformation solutions using IGEO, our in-house geoinformation platform.
Throughout my career, I have led the end-to-end development of product groups, provided technical consulting, and managed the implementation of GIS/IT solutions, primarily for local government, as well as infrastructure and telecom sectors.
I also spent a year intensively developing new functionalities and enhancing an internal billing system. A key challenge, tackled alongside my team, was optimizing performance to handle over 10,000 transactions per second within a globally distributed system while meeting diverse billing demands from internal users (Sales, AM, CSM).
My expertise has been instrumental in establishing the Central Integration Platform for the Croatian Meteorological and Hydrological Service, enabling seamless access to meteorological, hydrological, and air quality data through a unified interface.
Sessions
Handling "big data" (large volumes of data) primarily generated by automated weather stations and their installed sensors, requires special attention across the processes of data collection, processing/transformation, storage, presentation, and delivery.
The first step is identifying data sources and selecting appropriate access methods. Next comes data transformation, which depends on the data type and the relational database model (RDBMS) into which we plan to transfer it using adequate ETL tools. A critical aspect is ensuring the storage model accommodates heterogeneous sources while remaining efficient for access and delivery. The final step involves interfaces, either machine-based (API) or user-based (UI), that enable seamless data access and consumption.
Challenges increase significantly when dealing with spatial data, which comes in various forms, including numerical measurements, raster satellite imagery, radar images, and numerical forecasting models.
Each stage demands thorough planning and optimization to handle billion-record tables and terabytes of raster data effectively.
In this presentation, we will share our experiences in managing large datasets from diverse location and weather sensors, as well as spatial imagery, highlighting the complexities of processing and visualizing data.