Glue is a ETL, Extract Transport Load, service. It is serverless and used to transform and prepare data for analytics. For example you could take data for S3 and RDS and load it into Redshift. You can also convert data into the Parquet format utilizing glue, which is columnar, making it better suited for analysts that are run utilizing tables and columns. Glue also utilizes the AWS Glue Data Catalog, this send out Glue Crawlers to collect data from various data sources and write the metadata to AWS Glue Catalog, this data is then used for ETL by AWS Glue. Glue job bookmarks can be used to prevent reprocessing old data. Glue studio can be used to monitor Glue jobs through a visual dashboard as well. Another option of Glue is Databrew which allows you to clean and normalize data for simpler analysis, this can be used alone or alongside glue for additional analysis once the data is cleaned. Glue Elastic views takes data from multiple sources, and without custom code, builds virtual tables with SQL and analyzes changes in the source data to update the table this table can then be forwarded to another database for analytics