.png%3Falt%3Dmedia%26token%3Da5dfada4-1947-4d21-ac3f-2b58f9581a40)
.png%3Falt%3Dmedia%26token%3Da5dfada4-1947-4d21-ac3f-2b58f9581a40)
Application Performance
The next section of the Summary Dashboard displays a breakdown of your application’s performance. It builds upon the information displayed in the top section of the Dashboard. In this section, you are presented with two distinct charts:
- Current Impact Breakdown chart
- Worst Performers chart
This is a sunburst chart displaying a breakdown of the currently impacted areas of your application. This chart is dynamic, meaning it only displays SLOs that are being impacted at the present moment. Every outer circle of this chart breaks down the information provided in the inner circle before it.
In the inner most circle of the chart you are able to see which dimensions of the application are being impacted. The next circle of the chart breaks down each one of these dimensions into specific dimension instances. In this way, you are able to find out which specific instances within each dimension are actually being impacted at the present moment. The next circle breaks down each dimension instance into SLO categories. This information allows you to understand the type of issue that might be affecting their application, for example distinguish between availability and latency issues. The final and outermost circle in the chart breaks down each SLO category into the actual SLOs that are being impacted. Each SLO element in this circle displays the SLO's name and is colored following a heat map logic, where darker elements represent SLOs with hight levels of impact, whereas lighter ones represent slightly impacted SLOs. You can click any of the SLO elements of the outermost circle to access that SLO's Insights page.
Summary of sunburst chart layout (moving outwards from the circle's center):
- Level 1 (inner most circle): dimensions
- Level 2: dimension instances
- Level 3: SLO category
- Level 4: currently impacted SLOs
Current Impact Breakdown chart
To summarize, through this chart's layout you are able to understand which specific areas of your application are being impacted in the present moment, and correlate that impact to specific SLOs. This allows you to focus on the areas that you deem more important and know exactly what components of your application to target in order to solve the issues that are impacting your application.
This is a bar chart where you are able to see the top under performing dimension instances within your application. This enables you to identify areas of your application that are at risk of underperforming and to prioritize reliability efforts against new feature development. Each dimension instance is represented by a bar in this chart. The size of each bar is determined by the absolute number of SLOs within the group. Furthermore, each bar is divided into three sections: Non-Compliant, In Danger and Compliant. The size of each section is determined by the number of SLOs in that are in state that the section represents.
You can hover your mouse over any of the bar's sections to see a tooltip displaying the absolute number of SLOs that each section is representing and also the remaining error budget value for the worst SLO within that category.
Worst Performers chart
The display order of each dimension instance in the chart is determined by their performance. The performance of each dimension instance is determined by the relative number of non-compliant SLOs within the instance. Instances with a high percentage of non-compliant SLOs have a worst performance and as such appear first in the chart. If an instance has the same percentage of non-compliant SLOs as another instance, or no non-compliant SLOs at all, then their relative order is determined by the percentage of in danger SLOs. The same logic applies to the In Danger and Compliant SLO states.
Last modified 10mo ago