The geospatial data foundation of the COOLJAPAN scientific computing ecosystem just went fully cloud-native and production-ready.
Today we released OxiGDAL 0.1.3 — a complete, production-grade pure Rust geospatial data abstraction library designed as a modern replacement for GDAL.
No C. No C++. No Fortran. No PROJ/GEOS system dependencies.
No build hell. No 1 GB+ Docker images.
Just clean, memory-safe, blazing-fast geospatial processing that compiles to a single static binary (or <1 MB WASM) and runs everywhere — from laptops to browsers to embedded devices to cloud clusters.
Why OxiGDAL 0.1.3 is a game changer
For decades, geospatial workflows meant depending on the massive GDAL C++ library (with its complex toolchain, PROJ, GEOS, and 1 GB+ Docker images).
These tools are powerful but suffer from:
- C/C++ memory unsafety and segfaults
- Heavy system dependencies and installation nightmares
- Poor WASM/embedded/cloud-native support
- Slow Python bindings (GeoPandas)
- Massive binaries and vendor lock-in
OxiGDAL 0.1.3 ends all of that.
It delivers competitive or superior performance while being 100% memory-safe and portable.
Notable results:
- GeoParquet vs GeoPandas: 10× faster
- COG tile access (local SSD): < 10 ms
- COG tile access (S3/GCS): < 100 ms
- GeoTIFF metadata reading: < 5 ms
- PROJ batch transform (1 M points): < 10 ms
- Docker image size: < 50 MB (vs 1 GB+ for GDAL)
Technical Deep Dive: How We Rebuilt GDAL in Pure Rust
The architecture uses 69 workspace crates organized into clean functional layers, radically optimized for modern Rust and cloud-native use:
-
Core & Algorithms (
oxigdal-core,oxigdal-proj,oxigdal-algorithms)
211+ embedded EPSG definitions, 20+ map projections (including Japan Plane Rectangular), SIMD-accelerated raster/vector ops (AVX-512/NEON). -
Format Drivers (11 crates)
GeoTIFF/COG, GeoJSON (RFC 7946), GeoParquet (Arrow-native), Zarr v2/v3, FlatGeobuf, Shapefile, NetCDF, HDF5, GRIB1/2, JPEG2000 (EBCOT decoder), VRT. -
Cloud & Storage (
oxigdal-cloud)
Async S3/GCS/Azure Blob via HTTP range requests, pure-Rust compression (OxiARC integration), advanced caching. -
GPU & Enterprise
wgpu compute shaders, OGC WMS 1.3.0 / WFS 2.0.0 services, Raft HA clustering, Kafka/Kinesis streaming, AES-256-GCM security. -
Bindings & CLI
PyO3 (NumPy interop), WASM (<1 MB gzipped), Node.js, full CLI (oxigdal info,convert,warp,dem).
Key Rust advantages:
- Zero C/C++/Fortran dependencies in default features
- Fearless concurrency and guaranteed memory safety
- Pay-for-what-you-use modular features
- Trivial cross-compilation to WASM, iOS, Android, embedded
no_std
What’s inside 0.1.3 (released March 17)
- Stability improvements and CI hardening
- Expanded EPSG coverage and JPEG2000 EBCOT tier-1 decoder
- Deeper OxiARC pure-Rust compression integration
- Enhanced cloud I/O and GPU shader performance
- Production readiness confirmed through 7,486 passing tests across 69 crates
- ~480,000 lines of pure Rust — zero warnings, clippy::unwrap_used = “deny”
This is the foundation
OxiGDAL is now the official geospatial backend for the entire COOLJAPAN scientific stack:
- SciRS2 / NumRS2 — all raster/vector and spatial algorithms
- OxiBLAS — accelerated linear algebra for terrain analysis
- OxiARC / OxiCode — cloud-native compression and serialization
- Spintronics / OxiHuman — geospatial integration for simulations and avatars
- OxiMedia / ToRSh — satellite imagery pipelines and RAG
- Future integration with OxiLean for formally verified CRS transformations
Repository: https://github.com/cool-japan/oxigdal
Star the repo if you want high-performance geospatial computing without the GDAL toolchain headaches.
The era of “just install GDAL and pray the C++ doesn’t segfault” is over.
Pure Rust cloud-native geospatial is here — fast, safe, and sovereign.
— KitaSan at COOLJAPAN OÜ March 17, 2026