Google Cloud has recently expanded its customizable observability dashboards to over 10 services, including Google Kubernetes Engine (GKE), Compute Engine, Cloud Run, Cloud Functions, Cloud Storage, Dataproc, Dataflow, MySQL System Insights, and a few others.
Joy Wang, senior product manager, and Peizong Guo, software engineer at Google, discussed the observability dashboard capabilities in a blog post. By integrating features directly into each Google Cloud product, the new dashboard simplifies the observability process, eliminating the need to switch between tools for troubleshooting. The new dashboard offers free features, such as adding or removing charts, incorporating raw logs, modifying chart configurations, and creating alerts.
With Google Cloud's customizable dashboard, observability engineers can add insights for their specific needs. Engineers can choose relevant metrics and logs from specific instances or clusters to pinpoint the required data. Engineers can create alerts directly from the dashboard or use recommended alerts, then view alert charts in context as well. Furthermore, they can visualize metrics, logs, and alerts, combining telemetry from various services like load balancers, compute, and databases into a single dashboard to streamline troubleshooting.
By clicking the "customize" button on the top right, engineers can unlock the full functionality of the custom dashboard. This includes configuring the GKE Managed Prometheus dashboard to incorporate custom business or application metrics not found in the predefined dashboard.
Source: Introducing customizable observability dashboards
In an article discussing the top 10 observability trends for 2024, Middleware.io cited an Oracle study from 2023, highlighting the negative impact of increased data on decision-making. The study found that 78% of workers are overwhelmed by data overload from many sources. The article also mentioned that developers and engineers are seeking to simplify their workflows by using a single centralized platform, rather than dealing with multiple interfaces and investing in numerous software solutions.
We also came across the 6th annual Observability Pulse survey report by logz.io, which revealed a notable increase in the adoption of a single observability tool, rising from 16% of respondents in 2023 to 21% in 2024.
Among the other features, using the customizable dashboard, engineers can tailor observability insights for individual instances or apply changes across all clusters. To delve deeper into the application's performance, they need to click the magnifying glass icon on any chart. This launches the Metrics Explorer directly within the dashboard, allowing you to investigate and customize metrics without switching to a different tool.
In GKE Compute Engine and Cloud Run dashboards, engineers can now easily identify the root cause of metric fluctuations. The new event annotation feature automatically marks significant changes on charts, providing a visual correlation between metric changes and potential triggering events. This streamlines troubleshooting and helps to address issues within the applications.
To start customizing your observability dashboards, navigate to the dashboard for your desired Google Cloud service (e.g., Compute Engine, GKE, Cloud Run). Look for the pencil icon, which indicates a customizable dashboard. Then, use the intuitive drag-and-drop interface to arrange charts, graphs, and other visualizations to suit your specific monitoring needs. You can even create alerts directly within the dashboard, add alert charts for contextual insights, or view alerts in Cloud Monitoring.
Wang has also invited the users to provide their feedback. To do so, click on the feedback icon at the top right of the dashboard page.