Métricas do Spark

Por padrão, o Dataproc Serverless ativa a coleta de métricas do Spark disponíveis, a menos que você use as propriedades de coleta de métricas do Spark para desativar ou substituir a coleta de uma ou mais métricas do Spark.

Propriedades da coleta de métricas do Spark

É possível usar as propriedades listadas nesta seção para desativar ou substituir a coleta de uma ou mais métricas do Spark disponíveis.

Propriedade Descrição
spark.dataproc.driver.metrics Use para desativar ou substituir as métricas do driver do Spark.
spark.dataproc.executor.metrics Use para desativar ou substituir as métricas do executor do Spark.
spark.dataproc.system.metrics Use para desativar as métricas do sistema do Spark.

Exemplos da CLI gcloud:

  • Desative a coleta de métricas do driver do Spark:

    gcloud dataproc batches submit spark \
        --properties spark.dataproc.driver.metrics="" \
        --region=region \
        other args ...
    
  • Modificar a coleta de métricas do driver padrão do Spark para coletar apenas BlockManager:disk.diskSpaceUsed_MB e DAGScheduler:stage.failedStages métricas:

    gcloud dataproc batches submit spark \
        --properties=^~^spark.dataproc.driver.metrics="BlockManager:disk.diskSpaceUsed_MB,DAGScheduler:stage.failedStages" \
        --region=region \
        other args ...
    

Métricas do Spark disponíveis

O Dataproc sem servidor coleta as métricas do Spark listadas nesta seção, a menos que você use as propriedades de coleta de métricas do Spark para desativar ou substituir a coleta.

custom.googleapis.com/METRIC_EXPLORER_NAME

Métricas do driver do Spark

Métrica Nome do Metrics Explorer
BlockManager:disk.diskSpaceUsed_MB spark/driver/BlockManager/disk/diskSpaceUsed_MB
BlockManager:memory.maxMem_MB spark/driver/BlockManager/memory/maxMem_MB
BlockManager:memory.memUsed_MB spark/driver/BlockManager/memory/memUsed_MB
DAGScheduler:job.activeJobs spark/driver/DAGScheduler/job/activeJobs
DAGScheduler:job.allJobs spark/driver/DAGScheduler/job/allJobs
DAGScheduler:messageProcessingTime spark/driver/DAGScheduler/messageProcessingTime
DAGScheduler:stage.failedStages spark/driver/DAGScheduler/stage/failedStages
DAGScheduler:stage.runningStages spark/driver/DAGScheduler/stage/runningStages
DAGScheduler:stage.waitingStages spark/driver/DAGScheduler/stage/waitingStages

Métricas do executor do Spark

Métrica Nome do Metrics Explorer
ExecutorAllocationManager:executors.numberExecutorsDecommissionUnfinished spark/driver/ExecutorAllocationManager/executors/numberExecutorsDecommissionUnfinished
ExecutorAllocationManager:executors.numberExecutorsExitedUnexpectedly spark/driver/ExecutorAllocationManager/executors/numberExecutorsExitedUnexpectedly
ExecutorAllocationManager:executors.numberExecutorsGracefullyDecommissioned spark/driver/ExecutorAllocationManager/executors/numberExecutorsGracefullyDecommissioned
ExecutorAllocationManager:executors.numberExecutorsKilledByDriver spark/driver/ExecutorAllocationManager/executors/numberExecutorsKilledByDriver
LiveListenerBus:queue.executorManagement.listenerProcessingTime spark/driver/LiveListenerBus/queue/executorManagement/listenerProcessingTime
executor:bytesRead spark/executor/bytesRead
executor:bytesWritten spark/executor/bytesWritten
executor:cpuTime spark/executor/cpuTime
executor:diskBytesSpilled spark/executor/diskBytesSpilled
executor:jvmGCTime spark/executor/jvmGCTime
executor:memoryBytesSpilled spark/executor/memoryBytesSpilled
executor:recordsRead spark/executor/recordsRead
executor:recordsWritten spark/executor/recordsWritten
executor:runTime spark/executor/runTime
executor:shuffleFetchWaitTime spark/executor/shuffleFetchWaitTime
executor:shuffleRecordsRead spark/executor/shuffleRecordsRead
executor:shuffleRecordsWritten spark/executor/shuffleRecordsWritten
executor:shuffleRemoteBytesReadToDisk spark/executor/shuffleRemoteBytesReadToDisk
executor:shuffleWriteTime spark/executor/shuffleWriteTime
executor:succeededTasks spark/executor/succeededTasks
ExecutorMetrics:MajorGCTime spark/executor/ExecutorMetrics/MajorGCTime
ExecutorMetrics:MinorGCTime spark/executor/ExecutorMetrics/MinorGCTime

Métricas do sistema

Métrica Nome do Metrics Explorer
agent:uptime agente/tempo de atividade
cpu:utilization CPU/uso
disk:bytes_used disk/bytes_used
disk:percent_used disk/percent_used
memory:bytes_used memory/bytes_used
memory:percent_used memory/percent_used
network:tcp_connections network/tcp_connections

Ver métricas do Spark

Para conferir as métricas do lote, clique em um ID de lote na página Batches do Dataproc no console do Google Cloud para abrir a página Details, que mostra um gráfico de métricas para a carga de trabalho do lote na guia Monitoring.

Consulte o Cloud Monitoring do Dataproc para mais informações sobre como visualizar as métricas coletadas.