Performance Tuning# FlinkSQL Performance Guide Deployment configuration optimization Optimize resource configurations Recommend configuration to improve throughput and address data skew Improve the performance configuration of the Streaming Join Optimize Group Aggregate Enable MiniBatch to improve data throughput Scenarios How to enable miniBatch Enable LocalGlobal to address general data skew Scenarios Limits Verification Enable PartialFinal to address data skew for COUNT DISTINCT Scenarios Method to enable PartialFinal Verification Aggregate With FILTER Modifier Scenarios Original statement Optimized statement Best Practice of TopN TopN algorithms TopN optimization methods Perform no-ranking output optimization Increase the cache size of TopN Include a time field in the PARTITION BY clause Efficient deduplication Deduplicate Keep FirstRow Deduplicate Keep LastRow Efficient built-in functions Replacing user-defined functions (UDFs) with built-in functions Use single-character delimiters in the KEYVALUE function Best practices of LIKE operator Use regular functions (REGEXP) with caution