@InterfaceAudience.Public public class ScanMetrics extends ServerSideScanMetrics
The data can be passed to mapreduce framework or other systems. We use atomic longs so that one thread can increment, while another atomically resets to zero after the values are reported to hadoop's counters.
Some of these metrics are general for any client operation such as put However, there is no need for this. So they are defined under scan operation for now.
COUNT_OF_ROWS_FILTERED_KEY, COUNT_OF_ROWS_FILTERED_KEY_METRIC_NAME, COUNT_OF_ROWS_SCANNED_KEY, COUNT_OF_ROWS_SCANNED_KEY_METRIC_NAME, countOfRowsFiltered, countOfRowsScanned
Constructor and Description |
---|
ScanMetrics()
constructor
|
addToCounter, createCounter, getCounter, getMetricsMap, getMetricsMap, hasCounter, setCounter
public static final String RPC_CALLS_METRIC_NAME
public static final String REMOTE_RPC_CALLS_METRIC_NAME
public static final String MILLIS_BETWEEN_NEXTS_METRIC_NAME
public static final String NOT_SERVING_REGION_EXCEPTION_METRIC_NAME
public static final String BYTES_IN_RESULTS_METRIC_NAME
public static final String BYTES_IN_REMOTE_RESULTS_METRIC_NAME
public static final String REGIONS_SCANNED_METRIC_NAME
public static final String RPC_RETRIES_METRIC_NAME
public static final String REMOTE_RPC_RETRIES_METRIC_NAME
public final AtomicLong countOfRPCcalls
public final AtomicLong countOfRemoteRPCcalls
public final AtomicLong sumOfMillisSecBetweenNexts
public final AtomicLong countOfNSRE
public final AtomicLong countOfBytesInResults
public final AtomicLong countOfBytesInRemoteResults
public final AtomicLong countOfRegions
public final AtomicLong countOfRPCRetries
public final AtomicLong countOfRemoteRPCRetries
Copyright © 2007–2019 Cloudera. All rights reserved.