Show HN: Spice Cayenne – SQL acceleration built on Vortex
a day ago
- #Spice.ai
- #Vortex
- #data-acceleration
- Spice Cayenne is a next-generation data accelerator designed for high-scale and low-latency data lake workloads.
- It combines the Vortex columnar format with an embedded metadata engine for faster queries and lower memory usage.
- Spice Cayenne outperforms existing accelerators like DuckDB and SQLite in benchmarks.
- The architecture separates data storage (Vortex) and metadata management for optimized performance.
- Vortex offers significant speed improvements over Apache Parquet, including faster random access and scans.
- Cayenne's metadata layer uses SQLite for ACID-compliant operations, ensuring fast and consistent access.
- Benchmarks show Cayenne achieves better query performance and lower memory consumption than DuckDB.
- Getting started with Cayenne involves simple configuration in Spicepod.yml.
- Upcoming features include index support, improved snapshot bootstrapping, and additional metadata backends.
- Cayenne is currently in beta, with a roadmap targeting stable release for terabyte and petabyte-scale workloads.