From 6153eb520058ba71b1db1f4c72e589a2ce83e4f1 Mon Sep 17 00:00:00 2001
From: Cassie Kays <86622587+cassiekays@users.noreply.github.com>
Date: Wed, 30 Oct 2024 14:37:47 -0700
Subject: [PATCH 1/3] changed two AOAI alerts to visible with the proper
 thresholds

---
 services/CognitiveServices/accounts/alerts.yaml | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/services/CognitiveServices/accounts/alerts.yaml b/services/CognitiveServices/accounts/alerts.yaml
index 9d8a784db..9a28dcb26 100644
--- a/services/CognitiveServices/accounts/alerts.yaml
+++ b/services/CognitiveServices/accounts/alerts.yaml
@@ -343,18 +343,18 @@
   description: Utilization % for a provisoned-managed deployment, calculated as (PTUs consumed / PTUs deployed) x 100.
   type: Metric
   verified: false
-  visible: false
+  visible: true
   tags: manual-ck
   properties:
     metricName: AzureOpenAIProvisionedManagedUtilizationV2
     metricNamespace: Microsoft.CognitiveServices/accounts
-    severity: 0
+    severity: 2
     windowSize: PT5M
     evaluationFrequency: PT1M
     timeAggregation: Total
     operator: GreaterThan
     criterionType: StaticThresholdCriterion
-    threshold: 0.0
+    threshold: 80
   guid: 693a3b37-1e2a-42d1-aaed-b1f374276d1c
 - name: AzureOpenAIRequests
   description: Number of calls made to the Azure OpenAI API over a period of time.
@@ -377,18 +377,18 @@
   description: Recommended latency (responsiveness) measure for streaming requests.
   type: Metric
   verified: false
-  visible: false
+  visible: true
   tags: manual-ck
   properties:
     metricName: AzureOpenAITimeToResponse
     metricNamespace: Microsoft.CognitiveServices/accounts
-    severity: 0
+    severity: 2
     windowSize: PT5M
     evaluationFrequency: PT1M
     timeAggregation: Total
     operator: GreaterThan
     criterionType: StaticThresholdCriterion
-    threshold: 0.0
+    threshold: 200ms
   guid: 995cc12a-1887-4669-92c5-70a6ca8bfe70
 - name: BaselineEstimatorOverallReward
   description: Baseline Estimator Overall Reward.

From befb971dfe3f4830a2c1c9f3d64bf1ddda9cd370 Mon Sep 17 00:00:00 2001
From: Cassie Kays <86622587+cassiekays@users.noreply.github.com>
Date: Thu, 31 Oct 2024 11:33:19 -0700
Subject: [PATCH 2/3] changed PG verified to true; added token cache threshold
 : visible and verified

---
 services/CognitiveServices/accounts/alerts.yaml | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/services/CognitiveServices/accounts/alerts.yaml b/services/CognitiveServices/accounts/alerts.yaml
index 9a28dcb26..da9fa66c6 100644
--- a/services/CognitiveServices/accounts/alerts.yaml
+++ b/services/CognitiveServices/accounts/alerts.yaml
@@ -308,19 +308,19 @@
 - name: AzureOpenAIContextTokensCacheMatchRate
   description: Percentage of the prompt tokens hit the cache, avaiable for PTU-managed.
   type: Metric
-  verified: false
-  visible: false
+  verified: true
+  visible: true
   tags: manual-ck
   properties:
     metricName: AzureOpenAIContextTokensCacheMatchRate
     metricNamespace: Microsoft.CognitiveServices/accounts
-    severity: 0
+    severity: 2
     windowSize: PT5M
     evaluationFrequency: PT1M
     timeAggregation: Total
     operator: GreaterThan
     criterionType: StaticThresholdCriterion
-    threshold: 0.0
+    threshold: 75
   guid: 81f8369c-65bf-4194-bfd2-ffdfa2470577
 - name: AzureOpenAIProvisionedManagedUtilization
   description: Utilization % for a provisoned-managed deployment, calculated as (PTUs consumed / PTUs deployed) x 100.
@@ -342,7 +342,7 @@
 - name: AzureOpenAIProvisionedManagedUtilizationV2
   description: Utilization % for a provisoned-managed deployment, calculated as (PTUs consumed / PTUs deployed) x 100.
   type: Metric
-  verified: false
+  verified: true
   visible: true
   tags: manual-ck
   properties:
@@ -376,7 +376,7 @@
 - name: AzureOpenAITimeToResponse
   description: Recommended latency (responsiveness) measure for streaming requests.
   type: Metric
-  verified: false
+  verified: true
   visible: true
   tags: manual-ck
   properties:

From c77f3422ca0f9eb1e000ba0bd0ab6d4a90b5ea22 Mon Sep 17 00:00:00 2001
From: Cassie Kays <86622587+cassiekays@users.noreply.github.com>
Date: Thu, 31 Oct 2024 12:04:17 -0700
Subject: [PATCH 3/3] fixed text in threshold for time to respond metric

---
 services/CognitiveServices/accounts/alerts.yaml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/services/CognitiveServices/accounts/alerts.yaml b/services/CognitiveServices/accounts/alerts.yaml
index da9fa66c6..2105993ce 100644
--- a/services/CognitiveServices/accounts/alerts.yaml
+++ b/services/CognitiveServices/accounts/alerts.yaml
@@ -374,7 +374,7 @@
     threshold: 0.0
   guid: a1528d17-f288-46b1-b084-8b8fe3af90fa
 - name: AzureOpenAITimeToResponse
-  description: Recommended latency (responsiveness) measure for streaming requests.
+  description: Recommended latency (responsiveness) measure for streaming requests. Time in milliseconds.
   type: Metric
   verified: true
   visible: true
@@ -388,7 +388,7 @@
     timeAggregation: Total
     operator: GreaterThan
     criterionType: StaticThresholdCriterion
-    threshold: 200ms
+    threshold: 200
   guid: 995cc12a-1887-4669-92c5-70a6ca8bfe70
 - name: BaselineEstimatorOverallReward
   description: Baseline Estimator Overall Reward.