diff --git a/telemetry_tempest_plugin/config.py b/telemetry_tempest_plugin/config.py
index a478158..68e47ce 100644
--- a/telemetry_tempest_plugin/config.py
+++ b/telemetry_tempest_plugin/config.py
@@ -96,7 +96,15 @@
     cfg.IntOpt('prometheus_scrape_interval',
                default=15,
                help="Scrape interval configured for prometheus. This can "
-                    "be used in test cases to properly configure autoscaling")
+                    "be used in test cases to properly configure autoscaling"),
+    cfg.StrOpt('autoscaling_instance_grouping',
+               default='prefix',
+               choices=['prefix', 'metadata'],
+               help="How to group instances for autoscaling testing. "
+                    "'prefix' relies on the instances having a common string "
+                    "at the start of their name. 'metadata' is a new and "
+                    "prefered way of grouping since 2024.2 relying on "
+                    "metering.server_group instance metadata")
 ]
 
 telemetry_services_opts = [
diff --git a/telemetry_tempest_plugin/scenario/telemetry_integration_prometheus_gabbits/autoscaling.yaml b/telemetry_tempest_plugin/scenario/telemetry_integration_prometheus_gabbits/autoscaling.yaml
index b66ae40..158cbde 100644
--- a/telemetry_tempest_plugin/scenario/telemetry_integration_prometheus_gabbits/autoscaling.yaml
+++ b/telemetry_tempest_plugin/scenario/telemetry_integration_prometheus_gabbits/autoscaling.yaml
@@ -57,22 +57,6 @@
           $.servers[1].status: ACTIVE
           $.servers.`len`: 2
 
-    - name: check prometheus query for the servers count .
-      desc: Check the Prometheus metric for the existence of servers
-      url: $ENVIRON['PROMETHEUS_SERVICE_URL']/api/v1/query
-      verbose: all
-      method: POST
-      request_headers:
-          content-type: application/x-www-form-urlencoded
-      data:
-         query=ceilometer_cpu{resource_name=~"te-$ENVIRON['RESOURCE_PREFIX'].*"}
-      poll:
-          count: 300
-          delay: 1
-      status: 200
-      response_json_paths:
-          $.data.result.`len`: 2
-
     - name: check alarm cpu_alarm_high ALARM
       verbose: all
       desc: Check the aodh alarm and its state
diff --git a/telemetry_tempest_plugin/scenario/telemetry_integration_prometheus_gabbits/create_stack.json b/telemetry_tempest_plugin/scenario/telemetry_integration_prometheus_gabbits/create_stack.json
index 036e5fb..32a8219 100644
--- a/telemetry_tempest_plugin/scenario/telemetry_integration_prometheus_gabbits/create_stack.json
+++ b/telemetry_tempest_plugin/scenario/telemetry_integration_prometheus_gabbits/create_stack.json
@@ -54,7 +54,7 @@
                             }
                         }
                     ],
-                    "query": "(rate(ceilometer_cpu{resource_name=~'te-$ENVIRON['RESOURCE_PREFIX'].*'}[$ENVIRON['PROMETHEUS_RATE_DURATION']s])) * 100"
+                    "query": $ENVIRON["QUERY"]
                 }
             },
             "web_server_scaledown_policy": {
@@ -82,7 +82,7 @@
                             }
                         }
                     ],
-                    "query": "(rate(ceilometer_cpu{resource_name=~'te-$ENVIRON['RESOURCE_PREFIX'].*'}[$ENVIRON['PROMETHEUS_RATE_DURATION']s])) * 100"
+                    "query": $ENVIRON["QUERY"]
                 }
             }
         }
diff --git a/telemetry_tempest_plugin/scenario/test_telemetry_integration_prometheus.py b/telemetry_tempest_plugin/scenario/test_telemetry_integration_prometheus.py
index 9c13b68..122a3f9 100644
--- a/telemetry_tempest_plugin/scenario/test_telemetry_integration_prometheus.py
+++ b/telemetry_tempest_plugin/scenario/test_telemetry_integration_prometheus.py
@@ -104,6 +104,28 @@
 
         super(PrometheusGabbiTest, cls).resource_cleanup()
 
+    def _prep_query(self, prometheus_rate_duration, resource_prefix):
+        if config.CONF.telemetry.autoscaling_instance_grouping == "metadata":
+            query = ("\"(rate(ceilometer_cpu{{server_group=~'stack_id'}}"
+                     "[{}s])) * 100\"").format(prometheus_rate_duration)
+            metadata_query = '''
+            {{
+                "str_replace": {{
+                    "template": {},
+                    "params": {{
+                        "stack_id": {{ "get_param": "OS::stack_id" }}
+                    }}
+                }}
+            }}
+            '''.format(query)
+            return metadata_query
+
+        else:
+            prefix_query = '''
+            "(rate(ceilometer_cpu{{resource_name=~'te-{}.*'}}[{}s])) * 100"
+            '''.format(resource_prefix, prometheus_rate_duration)
+            return prefix_query
+
     def _prep_test(self, filename):
         auth = self.os_primary.auth_provider.get_auth()
         networks = self.os_primary.networks_client.list_networks(
@@ -115,6 +137,7 @@
         prometheus_rate_duration = (
             config.CONF.telemetry.ceilometer_polling_interval
             + config.CONF.telemetry.prometheus_scrape_interval)
+        query = self._prep_query(prometheus_rate_duration, resource_prefix)
         os.environ.update({
             "USER_TOKEN": auth[0],
             "AODH_THRESHOLD": str(config.CONF.telemetry.alarm_threshold),
@@ -136,6 +159,7 @@
             "RESOURCE_PREFIX": resource_prefix,
             "PROMETHEUS_RATE_DURATION": str(prometheus_rate_duration),
             "LOAD_LENGTH": str(prometheus_rate_duration * 2),
+            "QUERY": query,
         })