sharath1709 commented on code in PR #686:
URL:
https://github.com/apache/flink-kubernetes-operator/pull/686#discussion_r2286270024
##########
flink-autoscaler/src/main/java/org/apache/flink/autoscaler/metrics/ScalingMetrics.java:
##########
@@ -78,6 +88,45 @@ public static void computeDataRateMetrics(
}
}
+ private static Optional<Double> getObservedTpr(
+ Map<FlinkMetric, AggregatedMetric> flinkMetrics,
+ Map<ScalingMetric, Double> scalingMetrics,
+ double numRecordsInPerSecond,
+ Configuration conf) {
+
+ // If there are no incoming records we return infinity to allow scale
down
+ if (numRecordsInPerSecond == 0) {
+ return Optional.of(Double.POSITIVE_INFINITY);
+ }
+
+ // We only measure observed tpr when we are catching up, that is when
the lag is beyond the
+ // configured observe threshold
+ boolean catchingUp =
+ scalingMetrics.getOrDefault(ScalingMetric.LAG, 0.)
+ >=
conf.get(AutoScalerOptions.OBSERVE_TRUE_PROCESSING_RATE_LAG_THRESHOLD)
+ .toSeconds()
+ * numRecordsInPerSecond;
+ if (!catchingUp) {
+ return Optional.empty();
+ }
+
+ double observedTpr =
+ computeObservedTprWithBackpressure(
+ numRecordsInPerSecond,
+
flinkMetrics.get(FlinkMetric.BACKPRESSURE_TIME_PER_SEC).getAvg());
+
+ return Double.isNaN(observedTpr) ? Optional.empty() :
Optional.of(observedTpr);
+ }
+
+ public static double computeObservedTprWithBackpressure(
Review Comment:
Semantically, I agree that the true processing rate should be defined as the
maximum rate in the absence of backpressure. However, the ObservedTpr is
specifically used to estimate catch-up time. This is guaranteed to be
inaccurate, since the job is already experiencing backpressure and the actual
catch-up time will be higher than what the autoscaler computes. As a result,
the autoscaler fails to meet the SLA defined by job.autoscaler.catch-up.duration
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
