[exporter/elasticsearch] Profiling: Duplicate events if count > 1 (#40947)

#### Description
Due to nanosecond timestamp resolution, the probability of events having
a Stacktrace.count value > 1 is very low. For the rare cases where it
is, we can write multiple identical events. The compression in
Elasticsearch will handle this very well.

On the read/query side, things would be come less complex, and random
sampling could be applied (requirement is that every event has the same
"weight" aka count). Random sampling can reduce latency for queries.

#### Link to tracking issue
Fixes #40946

Author: rockdaboot

.chloggen/elasticsearchexporter-profiling-duplicate-events.yaml

@@ -0,0 +1,27 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: enhancement
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: elasticsearchexporter
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Duplicate profiling events with count values larger than 1
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [40946]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext: Having all events with count=1 enables random sampling on the read path.
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: [user]

exporter/elasticsearchexporter/internal/serializer/otelserializer/serializeprofiles/transform.go

@@ -140,12 +140,11 @@ func stackPayloads(dic pprofile.ProfilesDictionary, resource pcommon.Resource, s
 			t := sample.TimestampsUnixNano().At(j)
 			event.TimeStamp = newUnixTime64(t)
 
+			count := 1
 			if j < sample.Value().Len() {
-				event.Count = uint16(sample.Value().At(j))
-			} else {
-				event.Count = 1 // restore default
+				count = int(sample.Value().At(j))
 			}
-			if event.Count > 0 {
+			for range count {
 				stackPayload = append(stackPayload, StackPayload{
 					StackTraceEvent: event,
 				})
@@ -221,7 +220,7 @@ func stackTraceEvent(dic pprofile.ProfilesDictionary, traceID string, sample ppr
 		EcsVersion:   EcsVersion{V: EcsVersionString},
 		HostID:       hostMetadata[string(semconv.HostIDKey)],
 		StackTraceID: traceID,
-		Count:        1, // TODO: Check whether count can be dropped with nanosecond timestamps
+		Count:        1, // Elasticsearch v9.2+ doesn't read the count value any more.
 		Frequency:    frequency,
 	}
 

exporter/elasticsearchexporter/internal/serializer/otelserializer/serializeprofiles/transform_test.go

@@ -509,7 +509,16 @@ func TestStackPayloads(t *testing.T) {
 						TimeStamp:    1000000000,
 						StackTraceID: wantedTraceID,
 						Frequency:    20,
-						Count:        2,
+						Count:        1,
+					},
+				},
+				{
+					StackTraceEvent: StackTraceEvent{
+						EcsVersion:   EcsVersion{V: EcsVersionString},
+						TimeStamp:    1000000000,
+						StackTraceID: wantedTraceID,
+						Frequency:    20,
+						Count:        1,
 					},
 				},
 			},