Package com.linkedin.davinci.stats
package com.linkedin.davinci.stats
-
ClassesClassDescriptionAbstractVeniceAggVersionedStats<STATS,
STATS_REPORTER extends AbstractVeniceStatsReporter<STATS>> AbstractVeniceStatsReporter<STATS>This class provides aggregate initialization support for host level ingestion stats classHostLevelIngestionStatsThis class is an aggregate place that keeps stats objects for multiple stores and total stats for each region for AggKafkaConsumerService.The store level stats for blob transferThe store level stats or the total stats will be unpopulated because there is no easy and reliable way to aggregate gauge stats such as rt topic offset lag.Class that exposes stats related to blob transfersThis class is the reporting class for stats classBlobTransferStatsMetrics reporting logics are registered intoMetricsRepositoryhere and send out to external metrics collection/visualization system.This class contains stats for DIV.This class is the reporting class for stats classDIVStats.This class contains stats for stats on the storage node host level.This class contains stats for store ingestion.This class is the reporting class for stats classIngestionStats.IsolatedIngestionProcessStats is a metrics collecting class that aims to collect metrics from isolated ingestion process.This class provides the stats for Kafka consumer service per region or per store.MetadataUpdateStats records metrics related to storage metadata update viaMainIngestionStorageMetadataServiceThis class is used to track the thread pool stats for the state transitions of the participant.Class that exposes RocksDB memory consumption stats based on all properties that are made available in https://github.com/facebook/rocksdb/blob/master/include/rocksdb/db.h#L870 Properties exist on a per RockDB database basis (equivalent to a Venice partition).VeniceVersionedStats<STATS,STATS_REPORTER extends AbstractVeniceStatsReporter<STATS>> VeniceVersionedStatsReporter<STATS,STATS_REPORTER extends AbstractVeniceStatsReporter<STATS>> This class serves as a latency sensor for write path that contains two types of stats: Avg and Max.