Use the Query Details page to understand the low-level details of how a SQL query is processed through Cloudera Impala. The initial information on the page can help you tune the performance of some kinds of queries, primarily those involving joins. The more detailed information on the page is primarily for troubleshooting with the assistance of Cloudera Support; you might be asked to attach the contents of the page to a trouble ticket.
- or - to download a text version of the query detail.
- - to download a binary version of the query detail.
The Query Plan section can help you diagnose and tune performance issues with queries. This information is especially useful to understand performance issues with join queries, such as inefficient order of tables in the SQL statement, lack of table and column statistics, and the need for query hints to specify a more efficient join mechanism. You can also learn valuable information about how queries are processed for partitioned tables.
The information in this section corresponds to the output of the EXPLAIN statement for the Impala query. Each fragment shown in the query plan corresponds to a processing step that is performed by the central coordinator node or distributed across the nodes in the cluster.
The Query Info section reports the attributes of the query, start and end time, duration, and statistics about HDFS access.
You can hover over the query attributes for information about supported
predicate and values. For example:
The Query Fragments section reports detailed low-level statistics for each query plan fragment, involving physical aspects such as CPU utilization, disk I/O, and network traffic. This is the primary information that Cloudera Support might use to help troubleshoot performance issues and diagnose bugs. The fields and their names might change as Impala internal processing is enhanced.