Please join me on September 9th at 10 am Pacific Time for my webinar entitled "Columnar Store for PostgreSQL using cstore_fdw". This webinar will give you a comprehensive overview of cstore_fdw, an open source PostgreSQL extension from Citus Data that lets you store tables in columnar form. During this live webinar, we will discuss how columnar storage differs from row-based storage. We will demonstrate the difference between columnar and row-based storage by storing data in compressed form in a column in contiguous blocks on disk, as opposed to storing each row separately.
You will learn which use-cases are the best fit for cstore_fdw along with performance tuning best practices. We will take a deeper dive into the Optimized Row Columnar (ORC) storage format that cstore_fdw uses and discuss key features such as skip-list indexes. We will also look at practical examples of how to make effective use of cstore_fdw, such as using it to reduce the storage size of tables with JSONB columns over 6x. Finally, I will give a live demo of how to set up cstore_fdw tables yourself and demonstrate some of the latest and upcoming features. PostgreSQL 9.5 adds foreign table inheritance, which lets you create tables that are part row-based and part columnar and move data between the different partitions, giving you the best of both worlds.
You can learn more by visiting the Citus Data Webinars page or use the button below to register now.
Learn how columnar storage can significantly reduce storage size, memory footprint, and I/O utilization for online analytical processing (OLAP) workloads