Data load
Vertica provides many ways to read data. You can load data into the database from a variety of sources, optionally transforming it in various ways. You can read data in place in its original format using external tables. You can use streaming, and you can import data from other Vertica databases. See Common use cases for an introduction.
Most data-loading operations, including external tables, revolve around the COPY statement, which has many options. This book focuses on COPY-based reads (data load and external tables). Other data-loading options supported by Vertica are described elsewhere:
- Apache Kafka integration
- Apache Spark integration
-
Using the HCatalog Connector (HCatalog is part of HDFS)
- Database export and import
-
Clients: Batch inserts using JDBC prepared statements, Using batch inserts (ODBC)
-
Inserting individual rows and queried data into a table: INSERT/INSERT...SELECT
In this section
- Common use cases
- Introduction to the COPY statement
- Global and column-specific options
- Specifying where to load data from
- Partitioned data
- Data formats
- Complex types
- Schema evolution
- Handling Non-UTF-8 input
- Transforming data during loads
- Distributing a load
- Using transactions to stage a load
- Handling messy data
- Monitoring COPY loads and metrics
- Automatic load
- Using load scripts
- Troubleshooting data loads
- Working with external data