Impala Pool Metrics

In addition to these base metrics, many aggregate metrics are available. If an entity type has parents defined, you can formulate all possible aggregate metrics using the formula base_metric_across_parents.

In addition, metrics for aggregate totals can be formed by adding the prefix total_ to the front of the metric name.

Use the type-ahead feature in the Cloudera Manager chart browser to find the exact aggregate metric name, in case the plural form does not end in "s".

For example, the following metric names may be valid for Impala Pool:

  • queries_ingested_rate_across_clusters
  • total_queries_ingested_rate_across_clusters

Some metrics, such as alerts_rate, apply to nearly every metric context. Others only apply to a certain service or role.

For more information about metrics, see Cloudera Manager Metrics and Metric Aggregation.

Metric Name Description Unit Parents CDH Version
impala_query_admission_wait_rate The time from submission for admission to its completion. Called 'admission_wait' in searches. ms per second   CDH 5, CDH 6
impala_query_bytes_streamed_rate The total number of bytes sent between Impala Daemons while processing this query. Called 'bytes_streamed' in searches. bytes per second   CDH 5, CDH 6
impala_query_cm_cpu_milliseconds_rate impala.analysis.cm_cpu_milliseconds.description ms per second   CDH 5, CDH 6
impala_query_hdfs_bytes_read_rate The total number of bytes read from HDFS by this query. Called 'hdfs_bytes_read' in searches. bytes per second   CDH 5, CDH 6
impala_query_hdfs_bytes_written_rate The total number of bytes written to HDFS by this query. Called 'hdfs_bytes_written' in searches. bytes per second   CDH 5, CDH 6
impala_query_memory_accrual_rate The total accrued memory usage by the query. This is computed by multiplying the average aggregate memory usage of the query by the query's duration. Called 'memory_accrual' in searches. byte seconds per second   CDH 5, CDH 6
impala_query_memory_spilled_rate Amount of memory spilled to disk. Called 'memory_spilled' in searches. bytes per second   CDH 5, CDH 6
impala_query_query_duration_rate The duration of the query in milliseconds. Called 'query_duration' in searches. ms per second   CDH 5, CDH 6
impala_query_thread_cpu_time_rate The sum of the CPU time used by all threads of the query. Called 'thread_cpu_time' in searches. ms per second   CDH 5, CDH 6
queries_ingested_rate Impala queries ingested by the Service Monitor queries per second cluster, impala CDH 5, CDH 6
queries_oom_rate Impala queries for which memory consumption exceeded what was allowed queries per second cluster, impala CDH 5, CDH 6
queries_rejected_rate Impala queries rejected from admission, commonly due to the queue being full or insufficient memory queries per second cluster, impala CDH 5, CDH 6
queries_spilled_memory_rate Impala queries that spilled to disk queries per second cluster, impala CDH 5, CDH 6
queries_successful_rate Impala queries that ran to completion successfully queries per second cluster, impala CDH 5, CDH 6
queries_timed_out_rate Impala queries that timed out waiting in queue during admission queries per second cluster, impala CDH 5, CDH 6