AWS Redshift Integration
Monitor query performance, WLM queue metrics, cluster storage, and COPY/UNLOAD throughput. AI query regression detection catches slow query patterns from data skew or plan changes before analysts notice.
How It Works
Deploy CloudFormation Stack
Launch the TigerOps CloudFormation template to configure Metric Streams for the AWS/Redshift namespace and grant read access to the Redshift system tables for query-level metrics.
Enable Audit Logging
Enable Redshift audit logging to S3. TigerOps ingests connection, user activity, and user logs to provide query-level execution times, error rates, and user activity patterns.
Configure System Table Queries
TigerOps runs read-only queries against Redshift system tables (STL_QUERY, SVL_QLOG, STV_WLM_SERVICE_CLASS_STATE) to collect WLM queue metrics and slow query data.
Set Cluster Health Alerts
Configure CPU, disk usage, and WLM queue wait time alerts per cluster. TigerOps fires alerts before disk space exhaustion and when WLM queues experience long-duration stalls.
What You Get Out of the Box
Query Performance Tracking
Query execution time percentiles, queued vs. executing query counts, and slow query identification via STL_QUERY. TigerOps surfaces the top-N long-running queries and correlates them with CPU and I/O spikes.
WLM Queue Metrics
Per-service-class queue depth, wait time, and execution slot utilisation from STV_WLM_SERVICE_CLASS_STATE. TigerOps alerts when queue wait time exceeds your SLA and identifies which workloads are competing for slots.
Cluster Storage & Disk Utilisation
PercentageDiskSpaceUsed per cluster with growth rate trend analysis. TigerOps projects storage exhaustion time and alerts with enough lead time to vacuum tables, archive data, or resize the cluster.
COPY & UNLOAD Throughput
Data ingestion throughput via COPY commands and data export throughput via UNLOAD. TigerOps tracks COPY job duration and row counts so you can detect slowdowns in your ETL pipelines.
Node & Cluster Health
Per-node CPU, memory, and network utilisation for multi-node clusters. TigerOps identifies skewed compute or storage distribution across nodes that can cause query performance degradation.
AI Query Regression Detection
TigerOps baselines query execution time per query hash and alerts when a query pattern starts running significantly slower than its historical baseline — catching regressions from data skew or plan changes.
Redshift Monitoring Setup
Enable audit logging and create the TigerOps monitoring user in Redshift.
-- Create TigerOps monitoring user (read-only)
CREATE USER tigerops_monitor PASSWORD '${TIGEROPS_DB_PASSWORD}';
-- Grant access to system tables for metrics
GRANT SELECT ON ALL TABLES IN SCHEMA pg_catalog TO tigerops_monitor;
GRANT SELECT ON stl_query TO tigerops_monitor;
GRANT SELECT ON svl_qlog TO tigerops_monitor;
GRANT SELECT ON stv_wlm_service_class_state TO tigerops_monitor;
GRANT SELECT ON svv_table_info TO tigerops_monitor;
GRANT SELECT ON stl_load_errors TO tigerops_monitor;
-- Enable audit logging (run via AWS CLI)
-- aws redshift enable-logging \
-- --cluster-identifier my-redshift-cluster \
-- --bucket-name my-redshift-audit-logs \
-- --s3-key-prefix redshift-audit/
-- Deploy TigerOps Redshift monitoring stack
-- aws cloudformation deploy \
-- --template-url https://tigerops-cfn.s3.amazonaws.com/redshift-integration.yaml \
-- --stack-name tigerops-redshift \
-- --capabilities CAPABILITY_IAM \
-- --parameter-overrides \
-- TigerOpsApiKey=${TIGEROPS_API_KEY} \
-- ClusterIdentifier=my-redshift-cluster \
-- MonitoringUser=tigerops_monitor \
-- MonitoringPassword=${TIGEROPS_DB_PASSWORD} \
-- DiskUsageWarningPercent=75 \
-- WLMQueueWaitThresholdSeconds=60Common Questions
Does TigerOps support both Redshift Provisioned and Serverless?
Yes. TigerOps collects metrics from both Redshift Provisioned clusters (AWS/Redshift namespace) and Redshift Serverless (AWS/Redshift-Serverless namespace) via Metric Streams. Both are displayed in the same Redshift dashboard.
What permissions does TigerOps need to query Redshift system tables?
TigerOps requires a read-only database user with SELECT permissions on STL_QUERY, SVL_QLOG, STV_WLM_SERVICE_CLASS_STATE, and SVV_TABLE_INFO. The user does not need access to any business data tables.
How does TigerOps collect WLM queue metrics without constant querying?
TigerOps runs lightweight system table queries every 60 seconds using a dedicated database connection with minimal resource impact. WLM metrics are also available via CloudWatch for provisioned clusters, which TigerOps combines with system table data for a complete picture.
Can TigerOps alert on Redshift COPY job failures?
Yes. TigerOps monitors the STL_LOAD_ERRORS system table and alerts when a COPY command results in errors. The alert includes the error count, the error message sample, and the S3 source path of the failed load.
Does TigerOps support Redshift cross-cluster datasharing metrics?
TigerOps tracks DatashareObjectCount and datashare query metrics for clusters participating in Redshift data sharing. This helps you understand cross-cluster query load and identify producers causing excessive consumer cluster impact.
Deep Visibility Into Your Redshift Data Warehouse
Query performance, WLM queue metrics, and ETL throughput monitoring with AI regression detection. Connect in minutes.