Elasticsearch 7.13 introduces a range of enhancements, including the general availability of the frozen tier, which enables efficient searching of petabytes of data stored on low-cost object storage like Amazon S3, Microsoft Azure Storage, and Google Cloud Storage. The release also features significant performance improvements in aggregations, with up to a 92% increase in term aggregation speed under certain conditions, and the introduction of the combined_fields query for better multi-field search capabilities. Enhancements in supervised machine learning include transitioning data frame analytics and inference to general availability, allowing users to train and deploy models for tasks such as outlier detection, regression, and classification. Additional security measures have been implemented, such as audit ignore policies, improved metadata for API keys, and warnings for unsecured clusters. Elasticsearch 7.13 also introduces model aliases for easier management of trained models and provides tools to transform runtime fields into indexed fields for better query performance.