eG Monitoring
 

Measures reported by HiveMtConnStmtTest

Prepared statements are an optimization that allows to parse a query only once but execute it multiple times with different concrete values. Prepared statements have two phases for execution: preparation of the statement, and execution of the statement. These prepared statements will be stored in the prepared statement cache. If the prepared statements are not executed as and when the statements are prepared, then, it may be owing to reasons such as processing bottlenecks on the database server or inadequate resources (CPU, memory, etc) allocated for execution of the statements or the database node in itself may be poorly configured. In addition, if more statements are prepared and are stored in the prepared statement cache, the statements may be evicted from the cache if the cache is not configured appropriately. To constantly maintain a steady configuration of the database node and its components, it is important to keep a vigil on the execution of the prepared statements. The Hive Statements test helps administrators in this regard!

This test reports how well the prepared statements are cached and executed per second. Using this test, administrators can be alerted to inadequate sizing of the prepared statement cache and processing bottlenecks on the target Apache Hive.

Outputs of the test: One set of results for the target Apache Hive

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
totPre Indicates the number of prepared statements during the last measurement period. Number

Apache Hive supports prepared statements. Prepared statements are an optimization that allows to parse a query only once but execute it multiple times with different concrete values.

stmtExe Indicates the number of prepared statements executed during the last measurement period. Number

Any statement that uses at least one bind marker will need to be prepared. After which the statement can be executed by providing concrete values for each of its marker.

A low value for this measure is a cause of concern. This may be due to the processing bottlenecks on the target data warehouse or inadequate resource allocated to the target data warehouse or the inadequate resource allocated to the execution of the statement.

stmtCache Indicates the number of prepared statements cached during the last measurement period. Number

 

stmtExeAvg Indicates the average time of prepared statements executed during the last measurement period. Milliseconds

 

stmtPreAvg Indicates the average time of prepared statements during the last measurement period. Milliseconds