xuanyuanking commented on a change in pull request #30336:
URL: https://github.com/apache/spark/pull/30336#discussion_r523976903
##########
File path: core/src/main/scala/org/apache/spark/internal/config/UI.scala
##########
@@ -203,4 +203,15 @@ private[spark] object UI {
.stringConf
.transform(_.toUpperCase(Locale.ROOT))
.createWithDefault("LOCAL")
+
+ val ENABLED_STREAMING_UI_CUSTOM_METRIC_LIST =
+ ConfigBuilder("spark.ui.sql.streaming.enabledCustomMetricList")
+ .internal()
+ .doc("Configures a list of custom metrics on Structured Streaming UI,
which are enabled. " +
+ "The list contains the name of the custom metrics separated by comma.
In aggregation" +
+ "only sum used.")
+ .version("3.1.0")
+ .stringConf
+ .toSequence
+ .createWithDefault(Nil)
Review comment:
Do we need to add some common custom metrics here by default?
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/streaming/ui/StreamingQueryStatisticsPage.scala
##########
@@ -199,49 +201,99 @@ private[ui] class StreamingQueryStatisticsPage(parent:
StreamingQueryTab)
"records")
graphUIDataForNumRowsDroppedByWatermark.generateDataJs(jsCollector)
- // scalastyle:off
- <tr>
- <td style="vertical-align: middle;">
- <div style="width: 160px;">
- <div><strong>Aggregated Number Of Total State Rows
{SparkUIUtils.tooltip("Aggregated number of total state rows.",
"right")}</strong></div>
- </div>
- </td>
- <td
class={"aggregated-num-total-state-rows-timeline"}>{graphUIDataForNumberTotalRows.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-num-total-state-rows-histogram"}>{graphUIDataForNumberTotalRows.generateHistogramHtml(jsCollector)}</td>
- </tr>
- <tr>
- <td style="vertical-align: middle;">
- <div style="width: 160px;">
- <div><strong>Aggregated Number Of Updated State Rows
{SparkUIUtils.tooltip("Aggregated number of updated state rows.",
"right")}</strong></div>
- </div>
- </td>
- <td
class={"aggregated-num-updated-state-rows-timeline"}>{graphUIDataForNumberUpdatedRows.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-num-updated-state-rows-histogram"}>{graphUIDataForNumberUpdatedRows.generateHistogramHtml(jsCollector)}</td>
- </tr>
- <tr>
- <td style="vertical-align: middle;">
- <div style="width: 160px;">
- <div><strong>Aggregated State Memory Used In Bytes
{SparkUIUtils.tooltip("Aggregated state memory used in bytes.",
"right")}</strong></div>
- </div>
- </td>
- <td
class={"aggregated-state-memory-used-bytes-timeline"}>{graphUIDataForMemoryUsedBytes.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-state-memory-used-bytes-histogram"}>{graphUIDataForMemoryUsedBytes.generateHistogramHtml(jsCollector)}</td>
- </tr>
+ val result =
+ // scalastyle:off
<tr>
<td style="vertical-align: middle;">
<div style="width: 160px;">
- <div><strong>Aggregated Number Of State Rows Dropped By
Watermark {SparkUIUtils.tooltip("Aggregated number of state rows dropped by
watermark.", "right")}</strong></div>
+ <div><strong>Aggregated Number Of Total State Rows
{SparkUIUtils.tooltip("Aggregated number of total state rows.",
"right")}</strong></div>
</div>
</td>
- <td
class={"aggregated-num-state-rows-dropped-by-watermark-timeline"}>{graphUIDataForNumRowsDroppedByWatermark.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-num-state-rows-dropped-by-watermark-histogram"}>{graphUIDataForNumRowsDroppedByWatermark.generateHistogramHtml(jsCollector)}</td>
+ <td
class={"aggregated-num-total-state-rows-timeline"}>{graphUIDataForNumberTotalRows.generateTimelineHtml(jsCollector)}</td>
+ <td
class={"aggregated-num-total-state-rows-histogram"}>{graphUIDataForNumberTotalRows.generateHistogramHtml(jsCollector)}</td>
</tr>
- // scalastyle:on
+ <tr>
+ <td style="vertical-align: middle;">
+ <div style="width: 160px;">
+ <div><strong>Aggregated Number Of Updated State Rows
{SparkUIUtils.tooltip("Aggregated number of updated state rows.",
"right")}</strong></div>
+ </div>
+ </td>
+ <td
class={"aggregated-num-updated-state-rows-timeline"}>{graphUIDataForNumberUpdatedRows.generateTimelineHtml(jsCollector)}</td>
+ <td
class={"aggregated-num-updated-state-rows-histogram"}>{graphUIDataForNumberUpdatedRows.generateHistogramHtml(jsCollector)}</td>
+ </tr>
+ <tr>
+ <td style="vertical-align: middle;">
+ <div style="width: 160px;">
+ <div><strong>Aggregated State Memory Used In Bytes
{SparkUIUtils.tooltip("Aggregated state memory used in bytes.",
"right")}</strong></div>
+ </div>
+ </td>
+ <td
class={"aggregated-state-memory-used-bytes-timeline"}>{graphUIDataForMemoryUsedBytes.generateTimelineHtml(jsCollector)}</td>
+ <td
class={"aggregated-state-memory-used-bytes-histogram"}>{graphUIDataForMemoryUsedBytes.generateHistogramHtml(jsCollector)}</td>
+ </tr>
+ <tr>
+ <td style="vertical-align: middle;">
+ <div style="width: 160px;">
+ <div><strong>Aggregated Number Of State Rows Dropped By
Watermark {SparkUIUtils.tooltip("Aggregated number of state rows dropped by
watermark.", "right")}</strong></div>
+ </div>
+ </td>
+ <td
class={"aggregated-num-state-rows-dropped-by-watermark-timeline"}>{graphUIDataForNumRowsDroppedByWatermark.generateTimelineHtml(jsCollector)}</td>
+ <td
class={"aggregated-num-state-rows-dropped-by-watermark-histogram"}>{graphUIDataForNumRowsDroppedByWatermark.generateHistogramHtml(jsCollector)}</td>
+ </tr>
+ // scalastyle:on
+
+ result ++= generateAggregatedCustomMetrics(query, minBatchTime,
maxBatchTime, jsCollector)
+ result
} else {
new NodeBuffer()
}
}
+ def generateAggregatedCustomMetrics(
+ query: StreamingQueryUIData,
+ minBatchTime: Long,
+ maxBatchTime: Long,
+ jsCollector: JsCollector): NodeBuffer = {
+ val result: NodeBuffer = new NodeBuffer
+
+ // This is made sure on caller side but put it here to be defensive
+ require(query.lastProgress.stateOperators.nonEmpty)
+ val enabledCustomMetrics =
parent.parent.conf.get(ENABLED_STREAMING_UI_CUSTOM_METRIC_LIST)
+ logDebug(s"Enabled custom metrics: $enabledCustomMetrics")
+ query.lastProgress.stateOperators.head.customMetrics.keySet().asScala
+ .filter(enabledCustomMetrics.contains(_)).map { metricName =>
+ val data = query.recentProgress.map(p =>
(parseProgressTimestamp(p.timestamp),
+ p.stateOperators.map(_.customMetrics.get(metricName).toDouble).sum))
+ val max = data.maxBy(_._2)._2
+
+ val graphUIData =
+ new GraphUIData(
+ s"aggregated-$metricName-timeline",
+ s"aggregated-$metricName-histogram",
+ data,
+ minBatchTime,
+ maxBatchTime,
+ 0,
+ max,
+ "")
+ graphUIData.generateDataJs(jsCollector)
+
+ result ++=
+ // scalastyle:off
+ <tr>
+ <td style="vertical-align: middle;">
+ <div style="width: 240px;">
+ <div><strong>Aggregated Custom Metric {s"$metricName"}
{SparkUIUtils.tooltip("Custom metric.", "right")}</strong></div>
+ </div>
+ </td>
+ <td
class={s"aggregated-$metricName-timeline"}>{graphUIData.generateTimelineHtml(jsCollector)}</td>
+ <td
class={s"aggregated-$metricName-histogram"}>{graphUIData.generateHistogramHtml(jsCollector)}</td>
+ </tr>
+ // scalastyle:on
Review comment:
indent
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/streaming/ui/StreamingQueryStatisticsPage.scala
##########
@@ -199,49 +201,99 @@ private[ui] class StreamingQueryStatisticsPage(parent:
StreamingQueryTab)
"records")
graphUIDataForNumRowsDroppedByWatermark.generateDataJs(jsCollector)
- // scalastyle:off
- <tr>
- <td style="vertical-align: middle;">
- <div style="width: 160px;">
- <div><strong>Aggregated Number Of Total State Rows
{SparkUIUtils.tooltip("Aggregated number of total state rows.",
"right")}</strong></div>
- </div>
- </td>
- <td
class={"aggregated-num-total-state-rows-timeline"}>{graphUIDataForNumberTotalRows.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-num-total-state-rows-histogram"}>{graphUIDataForNumberTotalRows.generateHistogramHtml(jsCollector)}</td>
- </tr>
- <tr>
- <td style="vertical-align: middle;">
- <div style="width: 160px;">
- <div><strong>Aggregated Number Of Updated State Rows
{SparkUIUtils.tooltip("Aggregated number of updated state rows.",
"right")}</strong></div>
- </div>
- </td>
- <td
class={"aggregated-num-updated-state-rows-timeline"}>{graphUIDataForNumberUpdatedRows.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-num-updated-state-rows-histogram"}>{graphUIDataForNumberUpdatedRows.generateHistogramHtml(jsCollector)}</td>
- </tr>
- <tr>
- <td style="vertical-align: middle;">
- <div style="width: 160px;">
- <div><strong>Aggregated State Memory Used In Bytes
{SparkUIUtils.tooltip("Aggregated state memory used in bytes.",
"right")}</strong></div>
- </div>
- </td>
- <td
class={"aggregated-state-memory-used-bytes-timeline"}>{graphUIDataForMemoryUsedBytes.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-state-memory-used-bytes-histogram"}>{graphUIDataForMemoryUsedBytes.generateHistogramHtml(jsCollector)}</td>
- </tr>
+ val result =
+ // scalastyle:off
<tr>
<td style="vertical-align: middle;">
<div style="width: 160px;">
- <div><strong>Aggregated Number Of State Rows Dropped By
Watermark {SparkUIUtils.tooltip("Aggregated number of state rows dropped by
watermark.", "right")}</strong></div>
+ <div><strong>Aggregated Number Of Total State Rows
{SparkUIUtils.tooltip("Aggregated number of total state rows.",
"right")}</strong></div>
</div>
</td>
- <td
class={"aggregated-num-state-rows-dropped-by-watermark-timeline"}>{graphUIDataForNumRowsDroppedByWatermark.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-num-state-rows-dropped-by-watermark-histogram"}>{graphUIDataForNumRowsDroppedByWatermark.generateHistogramHtml(jsCollector)}</td>
+ <td
class={"aggregated-num-total-state-rows-timeline"}>{graphUIDataForNumberTotalRows.generateTimelineHtml(jsCollector)}</td>
+ <td
class={"aggregated-num-total-state-rows-histogram"}>{graphUIDataForNumberTotalRows.generateHistogramHtml(jsCollector)}</td>
</tr>
- // scalastyle:on
+ <tr>
+ <td style="vertical-align: middle;">
+ <div style="width: 160px;">
+ <div><strong>Aggregated Number Of Updated State Rows
{SparkUIUtils.tooltip("Aggregated number of updated state rows.",
"right")}</strong></div>
+ </div>
+ </td>
+ <td
class={"aggregated-num-updated-state-rows-timeline"}>{graphUIDataForNumberUpdatedRows.generateTimelineHtml(jsCollector)}</td>
+ <td
class={"aggregated-num-updated-state-rows-histogram"}>{graphUIDataForNumberUpdatedRows.generateHistogramHtml(jsCollector)}</td>
+ </tr>
+ <tr>
+ <td style="vertical-align: middle;">
+ <div style="width: 160px;">
+ <div><strong>Aggregated State Memory Used In Bytes
{SparkUIUtils.tooltip("Aggregated state memory used in bytes.",
"right")}</strong></div>
+ </div>
+ </td>
+ <td
class={"aggregated-state-memory-used-bytes-timeline"}>{graphUIDataForMemoryUsedBytes.generateTimelineHtml(jsCollector)}</td>
+ <td
class={"aggregated-state-memory-used-bytes-histogram"}>{graphUIDataForMemoryUsedBytes.generateHistogramHtml(jsCollector)}</td>
+ </tr>
+ <tr>
+ <td style="vertical-align: middle;">
+ <div style="width: 160px;">
+ <div><strong>Aggregated Number Of State Rows Dropped By
Watermark {SparkUIUtils.tooltip("Aggregated number of state rows dropped by
watermark.", "right")}</strong></div>
+ </div>
+ </td>
+ <td
class={"aggregated-num-state-rows-dropped-by-watermark-timeline"}>{graphUIDataForNumRowsDroppedByWatermark.generateTimelineHtml(jsCollector)}</td>
+ <td
class={"aggregated-num-state-rows-dropped-by-watermark-histogram"}>{graphUIDataForNumRowsDroppedByWatermark.generateHistogramHtml(jsCollector)}</td>
+ </tr>
+ // scalastyle:on
+
+ result ++= generateAggregatedCustomMetrics(query, minBatchTime,
maxBatchTime, jsCollector)
+ result
} else {
new NodeBuffer()
}
}
+ def generateAggregatedCustomMetrics(
+ query: StreamingQueryUIData,
+ minBatchTime: Long,
+ maxBatchTime: Long,
+ jsCollector: JsCollector): NodeBuffer = {
+ val result: NodeBuffer = new NodeBuffer
+
+ // This is made sure on caller side but put it here to be defensive
+ require(query.lastProgress.stateOperators.nonEmpty)
+ val enabledCustomMetrics =
parent.parent.conf.get(ENABLED_STREAMING_UI_CUSTOM_METRIC_LIST)
Review comment:
How about `val enabledCustomMetrics =
SQLConf.get.getConf(ENABLED_STREAMING_UI_CUSTOM_METRIC_LIST)`?
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/streaming/ui/StreamingQueryStatisticsPage.scala
##########
@@ -199,49 +201,99 @@ private[ui] class StreamingQueryStatisticsPage(parent:
StreamingQueryTab)
"records")
graphUIDataForNumRowsDroppedByWatermark.generateDataJs(jsCollector)
- // scalastyle:off
- <tr>
- <td style="vertical-align: middle;">
- <div style="width: 160px;">
- <div><strong>Aggregated Number Of Total State Rows
{SparkUIUtils.tooltip("Aggregated number of total state rows.",
"right")}</strong></div>
- </div>
- </td>
- <td
class={"aggregated-num-total-state-rows-timeline"}>{graphUIDataForNumberTotalRows.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-num-total-state-rows-histogram"}>{graphUIDataForNumberTotalRows.generateHistogramHtml(jsCollector)}</td>
- </tr>
- <tr>
- <td style="vertical-align: middle;">
- <div style="width: 160px;">
- <div><strong>Aggregated Number Of Updated State Rows
{SparkUIUtils.tooltip("Aggregated number of updated state rows.",
"right")}</strong></div>
- </div>
- </td>
- <td
class={"aggregated-num-updated-state-rows-timeline"}>{graphUIDataForNumberUpdatedRows.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-num-updated-state-rows-histogram"}>{graphUIDataForNumberUpdatedRows.generateHistogramHtml(jsCollector)}</td>
- </tr>
- <tr>
- <td style="vertical-align: middle;">
- <div style="width: 160px;">
- <div><strong>Aggregated State Memory Used In Bytes
{SparkUIUtils.tooltip("Aggregated state memory used in bytes.",
"right")}</strong></div>
- </div>
- </td>
- <td
class={"aggregated-state-memory-used-bytes-timeline"}>{graphUIDataForMemoryUsedBytes.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-state-memory-used-bytes-histogram"}>{graphUIDataForMemoryUsedBytes.generateHistogramHtml(jsCollector)}</td>
- </tr>
+ val result =
+ // scalastyle:off
<tr>
<td style="vertical-align: middle;">
<div style="width: 160px;">
- <div><strong>Aggregated Number Of State Rows Dropped By
Watermark {SparkUIUtils.tooltip("Aggregated number of state rows dropped by
watermark.", "right")}</strong></div>
+ <div><strong>Aggregated Number Of Total State Rows
{SparkUIUtils.tooltip("Aggregated number of total state rows.",
"right")}</strong></div>
</div>
</td>
- <td
class={"aggregated-num-state-rows-dropped-by-watermark-timeline"}>{graphUIDataForNumRowsDroppedByWatermark.generateTimelineHtml(jsCollector)}</td>
- <td
class={"aggregated-num-state-rows-dropped-by-watermark-histogram"}>{graphUIDataForNumRowsDroppedByWatermark.generateHistogramHtml(jsCollector)}</td>
+ <td
class={"aggregated-num-total-state-rows-timeline"}>{graphUIDataForNumberTotalRows.generateTimelineHtml(jsCollector)}</td>
+ <td
class={"aggregated-num-total-state-rows-histogram"}>{graphUIDataForNumberTotalRows.generateHistogramHtml(jsCollector)}</td>
</tr>
- // scalastyle:on
+ <tr>
Review comment:
indent: 2 extra space
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
