Parquet / GeoParquet
Apr 28,2026

GISBox is a one-stop 3D GIS data editing, conversion and publishing platform that supports editing in multiple GIS formats such as OSGB/GEOTIFF/RVT, converting to 3DTiles/Terrain and publishing.

Introduction

Parquet / GeoParquet is a cloud-native data format built on Apache Parquet, a columnar data format, extended to efficiently store and process geospatial data. GeoParquet adds metadata specifications for handling geospatial information and is designed to enable high-speed processing of large-scale spatial data in cloud environments and distributed processing platforms. It is primarily used in environments such as big data analytics, cloud GIS, and geospatial data lakes, and is recognized as a next-generation geospatial data format gaining attention within the Open Geospatial Consortium (OGC) community.

File Structure

GeoParquet is saved as a single .parquet file, but its internal structure consists of multiple logical elements. The main components are as follows:

Data Columns: Using Parquet's characteristic columnar storage, each attribute is stored column by column. This allows only the necessary columns to be read, enabling high-speed processing of large-scale data.
Geometry Column: GeoParquet defines a dedicated column for storing spatial information. Geometries are typically stored in WKB (Well-Known Binary) format, enabling efficient management of spatial objects such as points, lines, and polygons.
Geo Metadata (geo metadata): An essential element of GeoParquet, stored as metadata within the file. It defines information such as the coordinate reference system (CRS), geometry type, and bounding box, allowing GIS software and analysis tools to correctly interpret the data as spatial data.
Row Groups: Data is divided into multiple row groups for storage. This enables distributed processing and parallel reading, improving the processing performance of large-scale geospatial data.
Compression Encoding: Parquet supports compression methods such as Snappy, GZIP, and ZSTD, reducing data size while achieving high-speed reading and writing.

Pros

Cloud-native design: GeoParquet is optimized for cloud storage and data lake environments, with high compatibility with distributed processing engines and big data platforms.
High read performance: The columnar storage structure allows reading only the necessary attribute columns, enabling high-speed query processing even for large-scale data.
High compression efficiency: Efficient compression algorithms and columnar data structure can significantly reduce data size compared to traditional formats like Shapefile.
Affinity with big data analytics: Because it can directly interface with big data processing tools such as Apache Spark and cloud analytics platforms, it is suitable for large-scale geospatial analysis.
Open specification: GeoParquet is an open specification and is increasingly being used in many GIS tools and data analysis platforms.

Cons

Still limited GIS software support: As a relatively new format, it may not be directly supported by some traditional GIS software applications.
Difficult to edit directly by humans: Because it is a binary format, it is difficult to edit directly with a text editor, unlike CSV or GeoJSON.
Not suitable for real-time editing purposes: The data is optimized for large-scale batch processing and analysis, and may not be suitable for data management requiring frequent updates or edits.
Limitations in GIS-specific functionalities: Some GIS-specific database functions such as topology or advanced spatial indexes are not provided in the standard specification.

Application Scenario

GeoParquet is a format suitable for storing and analyzing large-scale geospatial data. It is widely used for spatial data management in cloud-based GIS systems and data lake environments, and is utilized for big data analysis of urban data, traffic data, environmental observation data, and more. Additionally, by combining it with distributed processing platforms, it can efficiently process enormous amounts of spatial data, drawing attention in fields such as smart cities, remote sensing analysis, and geographic big data analytics. Furthermore, its adoption as a standard data exchange format is advancing in cloud-native GIS and next-generation spatial data platforms.