Metrics

Alluxio Metrics

Cache Storage

Metric	Labels	Type	Component	Description
alluxio_cached_storage_bytes	-	gauge	worker	amount of the cached data
alluxio_cached_capacity_bytes	-	gauge	worker	configured maximum cache storage
alluxio_cached_evicted_data_bytes_total	-	counter	worker	amount of the evicted data
alluxio_data_cached_files	-	gauge	worker	number of files that are currently cached, both fully or partially, in a worker. Note that summing this metric over multiple workers in a cluster does not produce a total number of unique files cached, as files may be replicated on multiple workers.
alluxio_data_cached_pages	-	gauge	worker	number of pages being cached in a worker

Data Access

Metric	Labels	Type	Component	Description
alluxio_data_access_bytes	method=read/write	histogram	fuse, worker	aggregated all data access requests
alluxio_data_throughput_bytes_total	method=read/write	counter	worker	counter of data throughput of all data access
alluxio_meta_operation_total	op=getStatus/listStatus/...	counter	worker	counter of rpc calls of the meta operations
alluxio_cached_data_read_bytes_total	-	counter	worker	number of cache hits in page store
alluxio_missed_data_read_bytes_total	-	counter	worker	number of cache misses in page store
alluxio_ufs_data_access_bytes_total	method=read/write	counter	fuse, worker	amount of the ufs access
alluxio_fuse_concurrency	method=Fuse.Getattr/Fuse.readdir/...	gauge	fuse	record the realtime concurrency for fuse method
alluxio_s3_api_throughput_bytes	method=read/write/list	histogram	fuse, worker	histogram of s3 IO API throughput
alluxio_s3_api_call_processing	method=GetObject/ListObjects/...	gauge	fuse, worker	counter of the s3 API calls that are being processed
alluxio_multi_replica_read_from_workers	cluster_name,local_cluster,hot_read	counter	client	number of bytes read by a client from Alluxio workers when reading multi-replica files
alluxio_passive_cache_async_loaded_files	result=submitted/success/failure	counter	worker	number of files that are loaded by a worker triggered by passive cache

Distributed Load

Metric	Labels	Type	Component	Description
alluxio_distributed_load_job_scanned_total	-	counter	coordinator	counter of the inodes scanned in distributed load
alluxio_distributed_load_job_processed_total	-	counter	coordinator	counter of the non empty file copies loaded in distributed load
alluxio_distributed_load_job_skipped_total	-	counter	coordinator	counter of the inodes skipped in distributed load
alluxio_distributed_load_job_failure_total	reason, final_attempt, worker	counter	coordinator	counter of the distributed load failure
alluxio_distributed_load_job_dispatched_size_total	-	counter	coordinator	counter of the bytes dispatched in distributed load
alluxio_distributed_load_job_loaded_bytes_total	-	counter	coordinator	counter of the bytes loaded in distributed load

License

Metric	Labels	Type	Description
alluxio_license_expiration_date	-	gauge	the license expiration date in epoch time format

Process and JVM Metrics

Metric	Labels	Type	Description
process_start_time_seconds	-	gauge	start time of the process since unix epoch in seconds
process_cpu_seconds_total	-	counter	total user and system CPU time spent in seconds
jvm_threads_current	-	gauge	current thread count of a JVM
jvm_memory_used_bytes	area=heap/nonheap	gauge	used bytes of a given JVM memory area
jvm_memory_max_bytes	area=heap/nonheap	gauge	max (bytes) of a given JVM memory area
jvm_gc_collection_seconds	gc="G1 Young Generation"/"G1 Old Generation"/...	summary	time spent in a given JVM garbage collector in seconds