Monitoring Redshift

Amazon Redshift is a fully-managed, highly-scalable database service on the AWS cloud. It can host data in a Redshift cluster and provide an interface to a data lake. Redshift can scale from a few hundred gigabytes of data to more than a petabyte.

Redshift uses columnar storage, data compression, and zone maps to deliver fast parallel processing. It also integrates machine learning and results caching to manage large workloads and improve response time with repeated workloads.

Business Challenge

While basic tools exist to monitor Redshift, these tools do not show relationships between data in a meaningful way, nor can they provide historical context. Amazon CloudWatch provides host performance metrics while Redshift operational data is available in system views, however neither interface provides a complete picture of your Redshift cluster. For environments with large clusters, the picture becomes even less clear.

Foglight for Redshift delivers a comprehensive overview of your Redshift clusters, combining CloudWatch metrics and query/load performance data with Host metrics. Combining this data provides a complete picture of your Redshift environment in a single, unified interface. No other tool provides this level of availability and performance monitoring with historical analysis. Foglight for Redshift allows DBA’s to manage their environment in the same way they manage other legacy platforms. Foglight for Redshift provides the detailed information necessary to optimize query, table, and node performance.

Key Features

Foglight for Redshift brings true enterprise-class monitoring for Redshift to the industry’s leading crossplatform database management solution. Leveraging both Redshift system data and the CloudWatch API, Foglight for Redshift combines performance and monitoring data into an industry-leading monitoring platform.

View all monitored Redshift clusters with health status, connections, workload, and critical performance metrics consolidated in a single dashboard. Drill down to a summary of cluster data with detailed information on nodes and tables including the key performance metrics and workload data necessary to ensure optimum performance.

Foglight for Redshift also provides information on WLM Queue states, active transactions, and performance of top queries. Users can quickly identify queries with long execution times or those that return a large amount of data or use excessive system resources. Queries to an S3 data lake made through Redshift are included in the performance monitoring.

Foglight for Redshift is preconfigured with features including out-of-the-box rules to provide automated notification of availability and performance and reports to provide regular updates to engineering and management. Foglight proactively creates performance baselines and notifies users of abnormalities.

Rules can be configured to trigger on pre-defined thresholds or on behavior that is unusual for a given timeframe. Notifications and reports are easily customized and tuned for your environment and specific business needs.

Additional details will be provided shortly.