Extend the platform,
empower your team.
Get insights into Google Vertex AI service metrics.
ExtensionThis Dynatrace extension leverages data collected from the Google Operations API to constantly monitor health and performance of Google Vertex AI services.
View and analyze metrics that are specific to Google Vertex AI, like Accelerator memory utilization, Memory utilization, CPU utilization, and more.
Analyze Google AI Platform logs.
Google AI Platform metric and log ingestion requires advanced GCP integration.
Below is a complete list of the feature sets provided in this version. To ensure a good fit for your needs, individual feature sets can be activated and deactivated by your administrator during configuration.
Metric name | Metric key | Description | Unit |
---|---|---|---|
Accelerator duty cycle | cloud.gcp.aiplatform_googleapis_com.prediction.online.deployment_resource_pool.accelerator.duty_cycle | - | Percent |
Accelerator memory usage | cloud.gcp.aiplatform_googleapis_com.prediction.online.deployment_resource_pool.accelerator.memory.bytes_used | - | Byte |
CPU utilization | cloud.gcp.aiplatform_googleapis_com.prediction.online.deployment_resource_pool.cpu.utilization | - | Percent |
Memory usage | cloud.gcp.aiplatform_googleapis_com.prediction.online.deployment_resource_pool.memory.bytes_used | - | Byte |
Network bytes received | cloud.gcp.aiplatform_googleapis_com.prediction.online.deployment_resource_pool.network.received_bytes_count | - | Byte |
Network bytes sent | cloud.gcp.aiplatform_googleapis_com.prediction.online.deployment_resource_pool.network.sent_bytes_count | - | Byte |
Replica count | cloud.gcp.aiplatform_googleapis_com.prediction.online.deployment_resource_pool.replicas | - | Count |
Replica target | cloud.gcp.aiplatform_googleapis_com.prediction.online.deployment_resource_pool.target_replicas | - | Count |
Accelerator duty cycle | cloud.gcp.aiplatform_googleapis_com.prediction.online.accelerator.duty_cycle | - | Percent |
Accelerator memory usage | cloud.gcp.aiplatform_googleapis_com.prediction.online.accelerator.memory.bytes_used | - | Byte |
CPU utilization | cloud.gcp.aiplatform_googleapis_com.prediction.online.cpu.utilization | - | Percent |
Number of online prediction errors | cloud.gcp.aiplatform_googleapis_com.prediction.online.error_count | - | Count |
Memory usage | cloud.gcp.aiplatform_googleapis_com.prediction.online.memory.bytes_used | - | Byte |
Network bytes received | cloud.gcp.aiplatform_googleapis_com.prediction.online.network.received_bytes_count | - | Byte |
Network bytes sent | cloud.gcp.aiplatform_googleapis_com.prediction.online.network.sent_bytes_count | - | Byte |
Number of online predictions | cloud.gcp.aiplatform_googleapis_com.prediction.online.prediction_count | - | Count |
Prediction latencies | cloud.gcp.aiplatform_googleapis_com.prediction.online.prediction_latencies | - | MilliSecond |
Private endpoint prediction latencies | cloud.gcp.aiplatform_googleapis_com.prediction.online.private.prediction_latencies | - | MilliSecond |
Private endpoint response count | cloud.gcp.aiplatform_googleapis_com.prediction.online.private.response_count | - | Count |
Replica count | cloud.gcp.aiplatform_googleapis_com.prediction.online.replicas | - | Count |
Response count | cloud.gcp.aiplatform_googleapis_com.prediction.online.response_count | - | Count |
Replica target | cloud.gcp.aiplatform_googleapis_com.prediction.online.target_replicas | - | Count |
Executing PipelineJobs | cloud.gcp.aiplatform_googleapis_com.executing_vertexai_pipeline_jobs | - | Count |
Executing PipelineTasks | cloud.gcp.aiplatform_googleapis_com.executing_vertexai_pipeline_tasks | - | Count |
Generate content requests per minute per project per base model | cloud.gcp.aiplatform_googleapis_com.generate_content_requests_per_minute_per_project_per_base_model | - | Count |
Online prediction dedicated requests per base model version | cloud.gcp.aiplatform_googleapis_com.online_prediction_dedicated_requests_per_base_model_version | - | Count |
Online prediction dedicated tokens per minute per base model version | cloud.gcp.aiplatform_googleapis_com.online_prediction_dedicated_tokens_per_base_model_version | - | Count |
Online prediction requests per base model | cloud.gcp.aiplatform_googleapis_com.online_prediction_requests_per_base_model | - | Count |
Online prediction tokens per minute per base model | cloud.gcp.aiplatform_googleapis_com.online_prediction_tokens_per_minute_per_base_model | - | Count |
Generate content requests per minute per project per base model quota exceeded error | cloud.gcp.aiplatform_googleapis_com.quota.generate_content_requests_per_minute_per_project_per_base_model.exceeded | - | Count |
Generate content requests per minute per project per base model quota limit | cloud.gcp.aiplatform_googleapis_com.quota.generate_content_requests_per_minute_per_project_per_base_model.limit | - | Count |
Generate content requests per minute per project per base model quota usage | cloud.gcp.aiplatform_googleapis_com.quota.generate_content_requests_per_minute_per_project_per_base_model.usage | - | Count |
Online prediction dedicated requests per base model version quota exceeded error | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_dedicated_requests_per_base_model_version.exceeded | - | Count |
Online prediction dedicated requests per base model version quota limit | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_dedicated_requests_per_base_model_version.limit | - | Count |
Online prediction dedicated requests per base model version quota usage | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_dedicated_requests_per_base_model_version.usage | - | Count |
Online prediction dedicated tokens per minute per base model version quota exceeded error | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_dedicated_tokens_per_base_model_version.exceeded | - | Count |
Online prediction dedicated tokens per minute per base model version quota limit | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_dedicated_tokens_per_base_model_version.limit | - | Count |
Online prediction dedicated tokens per minute per base model version quota usage | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_dedicated_tokens_per_base_model_version.usage | - | Count |
Online prediction requests per base model quota exceeded | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_requests_per_base_model.exceeded | - | Count |
Online prediction requests per base model quota limit | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_requests_per_base_model.limit | - | Count |
Online prediction requests per base model quota usage | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_requests_per_base_model.usage | - | Count |
Online prediction tokens per minute per base model quota exceeded | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_tokens_per_minute_per_base_model.exceeded | - | Count |
Online prediction tokens per minute per base model quota limit | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_tokens_per_minute_per_base_model.limit | - | Count |
Online prediction tokens per minute per base model quota usage | cloud.gcp.aiplatform_googleapis_com.quota.online_prediction_tokens_per_minute_per_base_model.usage | - | Count |
Character count | cloud.gcp.aiplatform_googleapis_com.publisher.online_serving.character_count | - | Count |
Characters | cloud.gcp.aiplatform_googleapis_com.publisher.online_serving.characters | - | Count |
Character Throughput | cloud.gcp.aiplatform_googleapis_com.publisher.online_serving.consumed_throughput.count | - | Count |
First token latencies | cloud.gcp.aiplatform_googleapis_com.publisher.online_serving.first_token_latencies | - | MilliSecond |
Model invocation count | cloud.gcp.aiplatform_googleapis_com.publisher.online_serving.model_invocation_count | - | Count |
Model invocation latencies | cloud.gcp.aiplatform_googleapis_com.publisher.online_serving.model_invocation_latencies | - | MilliSecond |
Token count | cloud.gcp.aiplatform_googleapis_com.publisher.online_serving.token_count | - | Count |
Tokens | cloud.gcp.aiplatform_googleapis_com.publisher.online_serving.tokens | - | Count |
Metric name | Metric key | Description | Unit |
---|---|---|---|
PipelineJob duration | cloud.gcp.aiplatform_googleapis_com.pipelinejob.duration | - | Second |
Completed PipelineTasks | cloud.gcp.aiplatform_googleapis_com.pipelinejob.task_completed_count | - | Count |
Metric name | Metric key | Description | Unit |
---|---|---|---|
Request count | cloud.gcp.visionai_googleapis_com.platform.connected_service.request_count | - | Count |
Request latencies | cloud.gcp.visionai_googleapis_com.platform.connected_service.request_latencies | - | MilliSecond |
Prediction count | cloud.gcp.visionai_googleapis_com.platform.custom_model.predict_count | - | Count |
Prediction latencies | cloud.gcp.visionai_googleapis_com.platform.custom_model.predict_latencies | - | MilliSecond |
Uptime | cloud.gcp.visionai_googleapis_com.platform.instance.uptime | - | MilliSecond |
Received bytes | cloud.gcp.visionai_googleapis_com.stream.network.received_bytes_count | - | Byte |
Received packets | cloud.gcp.visionai_googleapis_com.stream.network.received_packets_count | - | Count |
Sent bytes | cloud.gcp.visionai_googleapis_com.stream.network.sent_bytes_count | - | Byte |
Sent packets | cloud.gcp.visionai_googleapis_com.stream.network.sent_packets_count | - | Count |
Metric name | Metric key | Description | Unit |
---|---|---|---|
Datapoint count | cloud.gcp.aiplatform_googleapis_com.matching_engine.stream_update.datapoint_count | - | Count |
Stream update latencies | cloud.gcp.aiplatform_googleapis_com.matching_engine.stream_update.latencies | - | MilliSecond |
Request count | cloud.gcp.aiplatform_googleapis_com.matching_engine.stream_update.request_count | - | Count |
CPU request utilization | cloud.gcp.aiplatform_googleapis_com.matching_engine.cpu.request_utilization | - | Percent |
Current replicas | cloud.gcp.aiplatform_googleapis_com.matching_engine.current_replicas | - | Count |
Current shards | cloud.gcp.aiplatform_googleapis_com.matching_engine.current_shards | - | Count |
Memory usage | cloud.gcp.aiplatform_googleapis_com.matching_engine.memory.used_bytes | - | Byte |
Request latency | cloud.gcp.aiplatform_googleapis_com.matching_engine.query.latencies | - | MilliSecond |
Request count | cloud.gcp.aiplatform_googleapis_com.matching_engine.query.request_count | - | Count |
Metric name | Metric key | Description | Unit |
---|---|---|---|
Request count | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.online_serving.request_count | - | Count |
Response bytes count | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.online_serving.serving_bytes_count | - | Count |
Request latency | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.online_serving.serving_latencies | - | MilliSecond |
Running sycs | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.running_sync | - | Count |
Serving data ages | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.serving_data_ages | - | Second |
Serving data by synced time | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.serving_data_by_sync_time | - | Count |
CPU load | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.storage.bigtable_cpu_load | - | Percent |
CPU load (hottest node) | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.storage.bigtable_cpu_load_hottest_node | - | Percent |
Node count | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.storage.bigtable_nodes | - | Count |
Optimized node count | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.storage.optimized_nodes | - | Count |
Bytes stored | cloud.gcp.aiplatform_googleapis_com.featureonlinestore.storage.stored_bytes | - | Byte |
CPU load | cloud.gcp.aiplatform_googleapis_com.featurestore.cpu_load | - | Percent |
CPU load (hottest node) | cloud.gcp.aiplatform_googleapis_com.featurestore.cpu_load_hottest_node | - | Percent |
Node count | cloud.gcp.aiplatform_googleapis_com.featurestore.node_count | - | Count |
Entities updated on the Featurestore online storage | cloud.gcp.aiplatform_googleapis_com.featurestore.online_entities_updated | - | Byte |
Latencies | cloud.gcp.aiplatform_googleapis_com.featurestore.online_serving.latencies | - | MilliSecond |
Request size | cloud.gcp.aiplatform_googleapis_com.featurestore.online_serving.request_bytes_count | - | Byte |
Serving count | cloud.gcp.aiplatform_googleapis_com.featurestore.online_serving.request_count | - | Count |
Response size | cloud.gcp.aiplatform_googleapis_com.featurestore.online_serving.response_size | - | Byte |
Billable bytes | cloud.gcp.aiplatform_googleapis_com.featurestore.storage.billable_processed_bytes | - | Byte |
Bytes stored | cloud.gcp.aiplatform_googleapis_com.featurestore.storage.stored_bytes | - | Byte |
Offline storage write for streaming write | cloud.gcp.aiplatform_googleapis_com.featurestore.streaming_write.offline_processed_count | - | Count |
Streaming write to offline storage delay time | cloud.gcp.aiplatform_googleapis_com.featurestore.streaming_write.offline_write_delays | - | Second |