Amazon Redshift announces the general availability of Multidimensional Data Layouts
Amazon Redshift announces the general availability of Multidimensional Data Layouts (MDDL) that dynamically sort data based on actual query filters, accelerating query performance. Unlike traditional sorting methods that sort data based on fixed columns, MDDL sorts data based on query filters (for example, Sales in US), achieving up to 10x better end-to-end performance compared to using only optimal single-column sort keys for query workloads with repetitive query filters.
For each table with an AUTO sort key, which is the default for tables without an explicit sort key, Redshift analyzes the table’s query history and automatically selects either a single-column sort key or MDDL for your table, depending on potential performance improvements for your workload. Redshift with MDDL automatically constructs a multidimensional virtual sort key that co-locates rows typically accessed by the same queries. This virtual column, equivalent to a new sort key for the table, is subsequently used during query execution to skip data blocks and even skip scanning entire predicate columns. Redshift with MDDL provides a more expressive generalization of existing compound and interleaved sort keys that significantly improves the performance of table scans, especially when your query workload contains repetitive query filters. For pre-existing tables with manually defined sort keys, you can take advantage of MDDL by altering the sortkey of the table to AUTO.
MDDL is available in all AWS commercial regions where Redshift is available. To get started, read the documentation, blog, and Amazon Science publication to learn more about benchmarks on query performance improvements.