Merge branch 'add-dropped-records-slos' into 'main'

Add dropped records SLO to UC2/UC4 (no Samza)

See merge request !310

theodolite-benchmarks/definitions/uc2-beam-flink/uc2-beam-flink-benchmark-operator.yaml

@@ -91,6 +91,18 @@ spec:
         threshold: 2000
         externalSloUrl: "http://localhost:80/evaluate-slope"
         warmup: 60 # in seconds
+    - name: droppedRecords
+      sloType: generic
+      prometheusUrl: "http://prometheus-operated:9090"
+      offset: 0
+      properties:
+        externalSloUrl: "http://localhost:8082"
+        promQLQuery: "sum(rate(flink_taskmanager_job_task_operator_org_apache_beam_runners_core_LateDataDroppingDoFnRunner_droppedDueToLateness[30s])) by (job) OR clamp_max(absent(notExists{job='absent'}),0)"
+        warmup: 60 # in seconds
+        queryAggregation: max
+        repetitionAggregation: median
+        operator: lte
+        threshold: 100
   kafkaConfig:
     bootstrapServer: "theodolite-kafka-kafka-bootstrap:9092"
     topics:

theodolite-benchmarks/definitions/uc2-flink/uc2-flink-benchmark-operator.yaml

@@ -91,6 +91,18 @@ spec:
         threshold: 2000
         externalSloUrl: "http://localhost:80/evaluate-slope"
         warmup: 60 # in seconds
+    - name: droppedRecords
+      sloType: generic
+      prometheusUrl: "http://prometheus-operated:9090"
+      offset: 0
+      properties:
+        externalSloUrl: "http://localhost:8082"
+        promQLQuery: "sum(rate(flink_taskmanager_job_task_operator_numLateRecordsDropped[30s])) by(job)"
+        warmup: 60 # in seconds
+        queryAggregation: max
+        repetitionAggregation: median
+        operator: lte
+        threshold: 100
   kafkaConfig:
     bootstrapServer: "theodolite-kafka-kafka-bootstrap:9092"
     topics:

theodolite-benchmarks/definitions/uc2-hazelcastjet/uc2-hazelcastjet-benchmark-operator.yaml

@@ -66,6 +66,18 @@ spec:
         threshold: 2000
         externalSloUrl: "http://localhost:80/evaluate-slope"
         warmup: 60 # in seconds
+    - name: droppedRecords
+      sloType: generic
+      prometheusUrl: "http://prometheus-operated:9090"
+      offset: 0
+      properties:
+        externalSloUrl: "http://localhost:8082"
+        promQLQuery: "sum(rate(com_hazelcast_jet_metrics_lateeventsdropped[30s])) by(job)"
+        warmup: 60 # in seconds
+        queryAggregation: max
+        repetitionAggregation: median
+        operator: lte
+        threshold: 100
   kafkaConfig:
     bootstrapServer: "theodolite-kafka-kafka-bootstrap:9092"
     topics:

theodolite-benchmarks/definitions/uc2-kstreams/uc2-kstreams-benchmark-operator.yaml

@@ -71,6 +71,18 @@ spec:
         threshold: 2000
         externalSloUrl: "http://localhost:80/evaluate-slope"
         warmup: 60 # in seconds
+    - name: droppedRecords
+      sloType: generic
+      prometheusUrl: "http://prometheus-operated:9090"
+      offset: 0
+      properties:
+        externalSloUrl: "http://localhost:8082"
+        promQLQuery: "sum(rate(kafka_streams_stream_task_metrics_dropped_records_total[30s])) by (job)"
+        warmup: 60 # in seconds
+        queryAggregation: max
+        repetitionAggregation: median
+        operator: lte
+        threshold: 100
   kafkaConfig:
     bootstrapServer: "theodolite-kafka-kafka-bootstrap:9092"
     topics:

theodolite-benchmarks/definitions/uc4-beam-flink/uc4-beam-flink-benchmark-operator.yaml

@@ -81,7 +81,7 @@ spec:
           resource: "uc4-load-generator-deployment.yaml"
           properties:
             loadGenMaxRecords: "150000"
-            numSensors: "4.0"
+            numSensors: "4"
   slos:
     - name: "lag trend"
       sloType: "lag trend"
@@ -92,6 +92,18 @@ spec:
         threshold: 2000
         externalSloUrl: "http://localhost:80/evaluate-slope"
         warmup: 60 # in seconds
+    - name: droppedRecords
+      sloType: generic
+      prometheusUrl: "http://prometheus-operated:9090"
+      offset: 0
+      properties:
+        externalSloUrl: "http://localhost:8082"
+        promQLQuery: "sum(rate(flink_taskmanager_job_task_operator_org_apache_beam_runners_core_LateDataDroppingDoFnRunner_droppedDueToLateness[30s])) by (job) OR clamp_max(absent(notExists{job='absent'}),0)"
+        warmup: 60 # in seconds
+        queryAggregation: max
+        repetitionAggregation: median
+        operator: lte
+        threshold: 100
   kafkaConfig:
     bootstrapServer: "theodolite-kafka-kafka-bootstrap:9092"
     topics:

theodolite-benchmarks/definitions/uc4-beam-samza/uc4-beam-samza-benchmark-operator.yaml

@@ -70,7 +70,7 @@ spec:
           resource: "uc4-load-generator-deployment.yaml"
           properties:
             loadGenMaxRecords: "150000"
-            numSensors: "4.0"
+            numSensors: "4"
   slos:
     - name: "lag trend"
       sloType: "lag trend"

theodolite-benchmarks/definitions/uc4-flink/uc4-flink-benchmark-operator.yaml

@@ -81,7 +81,7 @@ spec:
           resource: "uc4-load-generator-deployment.yaml"
           properties:
             loadGenMaxRecords: "150000"
-            numSensors: "4.0"
+            numSensors: "4"
   slos:
     - name: "lag trend"
       sloType: "lag trend"
@@ -92,6 +92,18 @@ spec:
         threshold: 2000
         externalSloUrl: "http://localhost:80/evaluate-slope"
         warmup: 60 # in seconds
+    - name: droppedRecords
+      sloType: generic
+      prometheusUrl: "http://prometheus-operated:9090"
+      offset: 0
+      properties:
+        externalSloUrl: "http://localhost:8082"
+        promQLQuery: "sum(rate(flink_taskmanager_job_task_operator_numLateRecordsDropped[30s])) by(job)"
+        warmup: 60 # in seconds
+        queryAggregation: max
+        repetitionAggregation: median
+        operator: lte
+        threshold: 100
   kafkaConfig:
     bootstrapServer: "theodolite-kafka-kafka-bootstrap:9092"
     topics:

theodolite-benchmarks/definitions/uc4-hazelcastjet/uc4-hazelcastjet-benchmark-operator.yaml

@@ -51,12 +51,12 @@ spec:
           resource: "uc4-load-generator-deployment.yaml"
           properties:
             container: "workload-generator"
-            variableName: "NUM_SENSORS"
+            variableName: "NUM_NESTED_GROUPS"
         - type: NumNestedGroupsLoadGeneratorReplicaPatcher
           resource: "uc4-load-generator-deployment.yaml"
           properties:
             loadGenMaxRecords: "150000"
-            numSensors: "4.0"
+            numSensors: "4"
   slos:
     - name: "lag trend"
       sloType: "lag trend"
@@ -67,6 +67,18 @@ spec:
         threshold: 2000
         externalSloUrl: "http://localhost:80/evaluate-slope"
         warmup: 60 # in seconds
+    - name: droppedRecords
+      sloType: generic
+      prometheusUrl: "http://prometheus-operated:9090"
+      offset: 0
+      properties:
+        externalSloUrl: "http://localhost:8082"
+        promQLQuery: "sum(rate(com_hazelcast_jet_metrics_lateeventsdropped[30s])) by(job)"
+        warmup: 60 # in seconds
+        queryAggregation: max
+        repetitionAggregation: median
+        operator: lte
+        threshold: 100
   kafkaConfig:
     bootstrapServer: "theodolite-kafka-kafka-bootstrap:9092"
     topics:

theodolite-benchmarks/definitions/uc4-kstreams/uc4-kstreams-benchmark-operator.yaml

@@ -56,12 +56,12 @@ spec:
           resource: "uc4-load-generator-deployment.yaml"
           properties:
             container: "workload-generator"
-            variableName: "NUM_SENSORS"
+            variableName: "NUM_NESTED_GROUPS"
         - type: NumNestedGroupsLoadGeneratorReplicaPatcher
           resource: "uc4-load-generator-deployment.yaml"
           properties:
             loadGenMaxRecords: "150000"
-            numSensors: "4.0"
+            numSensors: "4"
   slos:
     - name: "lag trend"
       sloType: "lag trend"
@@ -72,6 +72,18 @@ spec:
         threshold: 2000
         externalSloUrl: "http://localhost:80/evaluate-slope"
         warmup: 60 # in seconds
+    - name: droppedRecords
+      sloType: generic
+      prometheusUrl: "http://prometheus-operated:9090"
+      offset: 0
+      properties:
+        externalSloUrl: "http://localhost:8082"
+        promQLQuery: "sum(rate(kafka_streams_stream_task_metrics_dropped_records_total[30s])) by (job)"
+        warmup: 60 # in seconds
+        queryAggregation: max
+        repetitionAggregation: median
+        operator: lte
+        threshold: 100
   kafkaConfig:
     bootstrapServer: "theodolite-kafka-kafka-bootstrap:9092"
     topics: