Database Egress Optimization

6 posts · May 2022 – May 2026

What this trail covers

PlanetScale offers enhanced database egress optimization by integrating with DuckDB and supporting Change Data Capture (CDC) via Vitess's VStream API. This enables building data pipelines for analytics, data warehousing, and integration with other systems. The VStream API, exposed through VTGates, provides a unified change stream across sharded Vitess clusters, which can be consumed by tools like Debezium, Airbyte, and Fivetran, or directly via a gRPC client. This facilitates real-time data replication and enables building data pipelines for analytics, data warehousing, and integration with other systems.

All PlanetScale trails

May 2026Egress problems and where to find themThis post details strategies to minimize data transfer out of the database by optimizing query responses (fetching specific columns, using JSON filtering, pagination) and reducing request frequency (caching, CDNs). It also highlights the benefits of private connections (AWS PrivateLink, GCP Private Service Connect) for reducing egress costs and improving security. The post explains how to identify egress usage using `EXPLAIN` and PlanetScale Insights, and emphasizes the trade-offs between CPU overhead and reduced egress.
Dec 2025Using MotherDuck with PlanetScaleThis post introduces the integration of the `pg_duckdb` extension for PlanetScale Postgres databases, enabling the use of DuckDB for OLAP workloads. It details how to enable the extension, grant permissions, and connect to MotherDuck for offloading analytics queries. The post highlights the benefits of combining OLTP and OLAP processing and the ability to query external data sources like Parquet and Iceberg.
Jul 2024Building data pipelines with VitessThis post details how Vitess's VStream API, exposed through VTGates, provides a unified change stream across sharded Vitess clusters. It explains how this low-level gRPC primitive is used by popular CDC tools like Debezium, and by PlanetScale Connect for Airbyte and Fivetran sources. It also provides a practical example of using a direct gRPC client to stream changes from a Vitess cluster, demonstrating the output format and the process of setting up and running the example.
Nov 2023Announcing the Fivetran integrationThis post announces a new integration with Fivetran, enabling users to extract data from PlanetScale databases for ELT processes. It highlights the ease of setup using PlanetScale connection information and the benefits of Fivetran's ecosystem. The post also details how PlanetScale internally uses Fivetran for data warehousing and customer profiling, saving engineering time and enabling self-serve analytics.
Jun 2023Announcing the Hightouch integrationThis post details the integration of PlanetScale with Hightouch, a data syncing platform. It explains how PlanetScale databases can be used as both data sources and destinations within Hightouch workflows. The post provides examples of syncing data from PlanetScale to Google Sheets and vice-versa, highlighting the configuration steps for data sources, destinations, and sync methods. It also mentions the use of SQL queries or table scanning for data extraction and the ability to set automated or manual sync triggers.
Advertisement
May 2022Extract, load, and transform your data with PlanetScale ConnectThis post introduces PlanetScale Connect, a feature that enables ELT (Extract, Load, Transform) actions on PlanetScale data. It integrates with ELT platforms like Airbyte, allowing users to select their PlanetScale database as a source and connect to various destinations (e.g., Google BigQuery, AWS Redshift, Snowflake). The feature addresses schema discovery, initial data dumps, and incremental data synchronization. The benefits of ELT pipelines, such as offloading historical data, data consolidation, enrichment, productivity, and accuracy, are also discussed.