Google Vertex AI Metrics

Ship your Google Vertex AI Metrics via Telegraf to your Logit.io Stack

Follow the steps below to send your observability data to Logit.io

Metrics

Configure Telegraf to ship Google Vertex AI metrics to your Logit.io stacks via Logstash.

Install Integration

Please click on the Install Integration button to configure your stack for this source.

Set Credentials in GCP

@intro

Begin by heading over to the 'Project Selector' (opens in a new tab) and select the specific project from which you wish to send metrics.
- Progress to the 'Service Account Details' screen. Here, assign a distinct name to your service account and opt for 'Create and Continue'.
- In the 'Grant This Service Account Access to Project' screen, ensure the following roles: 'Compute Viewer', 'Monitoring Viewer', and 'Cloud Asset Viewer'.
- Upon completion of the above, click 'Done'.
- Now find and select your project in the 'Service Accounts for Project' list.
- Move to the 'KEYS' section.
- Navigate through Keys > Add Key > Create New Key, and specify 'JSON' as the key type.
- Lastly, click on 'Create', and make sure to save your new key.
Now add the environment variable for the key

On the machine run:
```
export GOOGLE_APPLICATION_CREDENTIALS=<your-gcp-key>
```

Install Telegraf

This integration allows you to configure a Telegraf agent to send your metrics, in multiple formats, to Logit.io.

Choose the installation method for your operating system:

When you paste the command below into Powershell it will download the Telegraf zip file. Once that is complete, press Enter again and the zip file will be extracted into C:\Program Files\InfluxData\telegraf\telegraf-1.31.2.

wget https://dl.influxdata.com/telegraf/releases/telegraf-1.31.2_windows_amd64.zip -UseBasicParsing -OutFile telegraf-1.31.2_windows_amd64.zip 
Expand-Archive .\telegraf-1.31.2_windows_amd64.zip -DestinationPath 'C:\Program Files\InfluxData\telegraf'

Configure the Telegraf input plugin

First you need to set up the input plug-in to enable Telegraf to scrape the GCP data from your hosts. This can be accomplished by incorporating the following code into your configuration file:

# Gather timeseries from Google Cloud Platform v3 monitoring API
[[inputs.stackdriver]]
  ## GCP Project
  project = "<your-project-name>"
 
  ## Include timeseries that start with the given metric type.
  metric_type_prefix_include = [
	"@metric_type",
  ]
 
  ## Most metrics are updated no more than once per minute; it is recommended
  ## to override the agent level interval with a value of 1m or greater.
  interval = "1m"

Read more about how to configure data scraping and configuration options for Stackdriver (opens in a new tab)

Configure the output plugin

Once you have generated the configuration file, you need to set up the output plug-in to allow Telegraf to transmit your data to Logit.io in Prometheus format. This can be accomplished by incorporating the following code into your configuration file:

[[outputs.http]]
  url = "https://@metricsUsername:@metricsPassword@@metrics_id-vm.logit.io:@vmAgentPort/api/v1/write"
  data_format = "prometheusremotewrite"
 
  [outputs.http.headers]
    Content-Type = "application/x-protobuf"
    Content-Encoding = "snappy"

Start Telegraf

From the location where Telegraf was installed (C:\Program Files\InfluxData\telegraf\telegraf-1.31.2) run the program providing the chosen configuration file as a parameter:

.\telegraf.exe --config telegraf-demo.conf

Once Telegraf is running you should see output similar to the following, which confirms the inputs, output and basic configuration the application has been started with: Powershell Telegraf information

View your metrics

Data should now have been sent to your Stack.

View My Data

If you don't see metrics take a look at How to diagnose no data in Stack below for how to diagnose common issues.

How to diagnose no data in Stack

If you don't see data appearing in your stack after following this integration, take a look at the troubleshooting guide for steps to diagnose and resolve the problem or contact our support team and we'll be happy to assist.

Telegraf Google Vertex AI Platform metrics Overview

Integrating Telegraf with Google Vertex AI allows organizations to monitor the performance of their AI models and the ML infrastructure. This includes tracking the usage metrics, operation statuses, and performance indicators of deployed models. Such metrics are crucial for ensuring the models perform as expected in production environments, providing insights into model accuracy, latency, and throughput.

However, the intricate nature and sheer volume of metrics generated by ML models and infrastructure pose a significant challenge for analysis and management. Logit.io offers a powerful solution to these challenges, providing a platform adept at handling the complex data landscape of AI and ML.

Logit.io enhances the ability to monitor and analyze ML model performance and infrastructure health, offering features like real-time alerting, comprehensive dashboards, and in-depth analytics. This enables organizations to quickly identify and address issues, optimize model performance, and make data-driven decisions to improve their AI initiatives.

For those utilizing Telegraf in conjunction with Google Vertex AI and seeking advanced analytics capabilities, Logit.io is an essential solution. Our platform streamlines the management of ML metrics, facilitating a deeper understanding of model behavior and operational efficiency. Effortlessly enhance your machine learning and AI workloads by synchronously sending crucial data from Google Vertex AI to Logit.io. This integration grants you full control and visibility over your systems, enabling you to monitor model performance, track data changes, and gain insights into system behaviour. Why not explore our comprehensive integration on Google Kubernetes Engine Logs to optimize your Kubernetes deployments, troubleshoot issues in real-time, and fine-tune cluster performance? Additionally, dive into Google Cloud GKE Metrics to gain a comprehensive understanding of your Kubernetes environment, allowing you to track resource utilization and monitor cluster health. You'll find that these integrations operate flawlessly within Logit.io's GCP logging (opens in a new tab) service.

Google Storage Transfer Metrics Google VM Manager Metrics