stream2000 commented on code in PR #9515: URL: https://github.com/apache/hudi/pull/9515#discussion_r1307089689
########## hudi-flink-datasource/hudi-flink/src/main/java/org/apache/hudi/metrics/FlinkWriteMetrics.java: ########## @@ -0,0 +1,130 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.hudi.metrics; + +import org.apache.hudi.common.model.HoodieCommitMetadata; +import org.apache.hudi.common.table.timeline.HoodieInstantTimeGenerator; +import org.apache.hudi.common.util.Option; +import org.apache.hudi.common.util.collection.Pair; + +import org.apache.flink.metrics.MetricGroup; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import java.text.ParseException; + +/** + * Common flink write commit metadata metrics + */ +public class FlinkWriteMetrics extends HoodieFlinkMetrics { + + private static final Logger LOG = LoggerFactory.getLogger(FlinkWriteMetrics.class); + + protected final String actionType; + + private long totalPartitionsWritten; + private long totalFilesInsert; + private long totalFilesUpdate; + private long totalRecordsWritten; + private long totalUpdateRecordsWritten; + private long totalInsertRecordsWritten; + private long totalBytesWritten; + private long totalScanTime; + private long totalCreateTime; + private long totalUpsertTime; + private long totalCompactedRecordsUpdated; + private long totalLogFilesCompacted; + private long totalLogFilesSize; + private long commitLatencyInMs; + private long commitFreshnessInMs; + private long commitEpochTimeInMs; + private long durationInMs; + + public FlinkWriteMetrics(MetricGroup metricGroup, String actionType) { + super(metricGroup); + this.actionType = actionType; + } + + @Override + public void registerMetrics() { + // register commit gauge + metricGroup.gauge(getMetricsName(actionType, "totalPartitionsWritten"), () -> totalPartitionsWritten); + metricGroup.gauge(getMetricsName(actionType, "totalFilesInsert"), () -> totalFilesInsert); + metricGroup.gauge(getMetricsName(actionType, "totalFilesUpdate"), () -> totalFilesUpdate); + metricGroup.gauge(getMetricsName(actionType, "totalRecordsWritten"), () -> totalRecordsWritten); + metricGroup.gauge(getMetricsName(actionType, "totalUpdateRecordsWritten"), () -> totalUpdateRecordsWritten); + metricGroup.gauge(getMetricsName(actionType, "totalInsertRecordsWritten"), () -> totalInsertRecordsWritten); + metricGroup.gauge(getMetricsName(actionType, "totalBytesWritten"), () -> totalBytesWritten); + metricGroup.gauge(getMetricsName(actionType, "totalScanTime"), () -> totalScanTime); + metricGroup.gauge(getMetricsName(actionType, "totalCreateTime"), () -> totalCreateTime); + metricGroup.gauge(getMetricsName(actionType, "totalUpsertTime"), () -> totalUpsertTime); + metricGroup.gauge(getMetricsName(actionType, "totalCompactedRecordsUpdated"), () -> totalCompactedRecordsUpdated); Review Comment: > Can this class be reused by Flink? Flink has its own metrics system and it seems hard to reuse hudi-spark's metrics system. For example, flink metrics core manages the `MetricsReporter` itself and we can config metrics reporters through flink config. However for hudi spark, we manager the reporter in hudi instead of Spark. We do can reuse the metrics defined in `hudi-client-common` for flink, however in this way we will bypass Flink's metrics system which may make hudi-flink metrics different from other flink metrics. I think in hudi-flink, we'd better use flink's metrics system so that those who already have a metrics for flink can easily integrate hudi-flink metrics. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: commits-unsubscr...@hudi.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
