Github user tdas commented on a diff in the pull request:
https://github.com/apache/spark/pull/15307#discussion_r81628507
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/StreamMetrics.scala
---
@@ -0,0 +1,252 @@
+/*
+ * 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.spark.sql.execution.streaming
+
+import scala.collection.mutable
+
+import com.codahale.metrics.{Gauge, MetricRegistry}
+
+import org.apache.spark.internal.Logging
+import org.apache.spark.metrics.source.{Source => CodahaleSource}
+import org.apache.spark.util.Clock
+
+class StreamMetrics(sources: Set[Source], triggerClock: Clock,
codahaleSourceName: String)
+ extends CodahaleSource with Logging {
+
+ import StreamMetrics._
+
+ // Trigger infos
+ private val triggerInfo = new mutable.HashMap[String, String]
+ private val sourceTriggerInfo = new mutable.HashMap[Source,
mutable.HashMap[String, String]]
+
+ // Rate estimators for sources and sinks
+ private val inputRates = new mutable.HashMap[Source, RateCalculator]
+ private val processingRates = new mutable.HashMap[Source, RateCalculator]
+ private val outputRate = new RateCalculator
+
+ // Number of input rows in the current trigger
+ private val numInputRows = new mutable.HashMap[Source, Long]
+ private var numOutputRows: Option[Long] = None
+ private var currentTriggerStartTimestamp: Long = -1
+ private var previousTriggerStartTimestamp: Long = -1
+ private var latency: Option[Double] = None
+
+ override val sourceName: String = codahaleSourceName
+ override val metricRegistry: MetricRegistry = new MetricRegistry
+
+ // =========== Initialization ===========
+
+ // Metric names should not have . in them, so that all the metrics of a
query are identified
+ // together in Ganglia as a single metric group
--- End diff --
Metrics are generally agnostic to sinks, but different sinks probably have
different ways of representing data visually. I tested with Ganglia (assuming
that its the most common one in production), found that it clusters based on
the last period. So multiple metrics names A.B.X, A.B.Y, A.B.Z will be put in a
cluster A.B .
Regarding validity, we using Codahales `Metrics.name(partialName1,
partialName2, ...)` underneath which is supposed to check formats and all that
stuff. So I am assuming that whatever passes through that check will be valid
for all sinks. And hyphen seems to be fine with Ganglia sink.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
