// the find
awslabs/amazon-redshift-utils
Amazon Redshift Utils contains utilities, scripts and view which are useful in a Redshift environment
A grab-bag of scripts, SQL views, and Python utilities from AWS for operating and tuning Redshift clusters. Covers everything from column encoding optimization to workload replay to cross-cluster data migration. If you run Redshift in production and have ever wanted a pre-built SQL view to check WLM queue state or lock contention, this is where those live.
The AdminViews collection is genuinely useful — views like v_generate_tbl_ddl and v_check_transaction_locks save hours of spelunking through Redshift system tables. The AnalyzeVacuumUtility actually understands when VACUUM is necessary by checking stl_alert_event_log rather than blindly running on all tables. SimpleReplay is non-trivial: it captures real workloads from audit logs and can replay them at different concurrencies, which is exactly what you need before resizing a cluster. The Lambda automation module (RedshiftAutomation) gives you CloudWatch-scheduled maintenance without standing up any persistent infrastructure.
This is a collection, not a product — quality and maintenance vary wildly by subdirectory, and several modules (SnapshotManager, older Lambda zips) are explicitly deprecated with no cleanup. The auth story is a known liability: KMS-encrypted passwords stored in config files, with the README itself calling it not the most secure solution. The IDC and Lake Formation migration utilities landed recently and have minimal test coverage; the integration tests require specific AWS environment setup with no mocking path, making CI on forks basically impossible. If you're not already deep in the Redshift ecosystem, the lack of cross-module cohesion means you'll spend time figuring out which tool solves your specific problem.