Documentation forSolarWinds Observability SaaS

Service Level Objective (SLO)

The Service Level Objectives (SLOs) help you define and monitor reliability targets for monitored entities. Each service or system has a different aspect when defining its health. SLOs provide an indication of reliability and trends, allowing users to make necessary changes.

Key aspects of SLO

Service Level Objectives (SLO)

SLOs define specific reliability targets for your service. They answer questions such as:

  • What uptime percentage should our service achieve?

  • How responsive should our API endpoints be?

  • What error rate is acceptable?

Service Level Indicators (SLI)

SLIs are the measurable metrics that gauge the health of your service. SLIs serve as the building blocks for SLOs. Examples of SLIs include:

  • Uptime percentage: The proportion of time your service is available.

  • Latency: The time taken to respond to requests.

  • Error rate: The frequency of errors encountered by users.

Error Budgets

An error budget represents the permissible deviation from your SLOs. It quantifies how much unreliability your service can tolerate while still meeting its commitments. If you exhaust your error budget, you should focus on reliability improvements.

Create SLOs

Before creating an SLO, make sure the entity and metrics you want to use are available in SolarWinds Observability SaaS.

  1. In SolarWinds Observability SaaS, click Settings > SLO > Create SLO.

  2. In the SLO Name field, type the name. The name will appear in dashboards and chart headers. SolarWinds recommends you give your SLO a clear name.

  3. In the Select a Scope drop-down, select either Single Entity if you're monitoring a database or a host, or Entity Group if you're monitoring a tag or a cluster.

  4. In the Entity Type drop-down, select the resource you want to monitor. Click Next.

  5. Define an objective.

    An SLO consists of the target percentage and the SLI metric you are tracking using ratio-based or threshold-based definitions. When you set a target for a metric-based SLO, the target percentage specifies what portion of the total events specified in the denominator of the SLO should be good events, while the time window specifies the rolling time period over which the target should be tracked.

    1. In the Name field, type the objective name.

    2. In the Indicator Type drop-down, select Threshold, Ratio Good, or Ratio Bad and type the percentage in the Target Percentage.

      • Threshold: Evaluates a metric against a defined threshold.

        1. In the Metric drop-down, select the threshold metric. These metrics focus on a specific value, such as the response time or error rate, and compare the value against a predefined threshold.

        2. Select the Aggregation Method.

        3. In Threshold Operator, select the conditional operator used to compare the SLI against the value.

        4. Specify the Threshold Value.

        5. (Optional) Add tag and click Add Objective to define another objective if necessary.

      • Ratio Good: Measures the percentage of good events compared with total events.

        1. Under Good Events, in the Metric drop-down, select the ratio metric. These metrics are used when the SLI is based on a ratio of good events to total events.

        2. Select the Aggregation Method.

        3. (Optional) Add tags.

        4. Under Total Events, in the Metric drop-down, select the ratio metric.

        5. (Optional) Add tags and click Add Objective to define another objective if necessary.

      • Ratio Bad: Measures the percentage of bad events compared with total events.

        1. Under Bad Events, in the Metric drop-down, select the ratio metric. These metrics are used when the SLI is based on a ratio of bad events to total events.

        2. Select the Aggregation Method.

        3. (Optional) Add tags.

        4. Under Total Events, in the Metric drop-down, select the ratio metric.

        5. (Optional) Add tags and click Add Objective to define another objective if necessary.

    3. Click Preview to view the Objective you defined. Click Next.

  6. On the Summary, review your configuration and click Create.

Review the Change History

After your SLOs are created, you can review their configuration and change history.

  1. In SolarWinds Observability SaaS, click Settings > SLO.

  2. Click the SLO you want to review.

  3. On the Configuration tab, review the scope, metrics, and targets defined for the SLI.

  4. Click the Change History tab to identify the object that changed, when it changed, and the user who made the change.

View SLO data in SolarWinds Observability SaaS

View data in the Service Levels list view

The Service Levels list view provides an overview of the SLOs configured in your current SolarWinds Observability SaaS environment. It gives you a central place to review SLO targets, associated entities, status, and performance information. Use the list view to:

  • Search for a specific SLO or associated entity.

  • Filter the list to focus on relevant SLOs.

  • Review the overall state of your SLOs.

  • Identify SLOs that may require further investigation.

  • Compare key SLO information from a single view.

The list includes the SLOs available in the current environment and displays their latest available information.

To open the Service Levels list view, click Service Levels in the left pane of SolarWinds Observability SaaS. Click Create SLO to add a new objective.

View SLO data in the Entity Explorer

  1. In the left pane of SolarWinds Observability SaaS, click Explore.

  2. Click the name of the entity you want to review the SLO/SLI data for.

  3. On the Entity Details page, click the SLO tab.

The following widgets are available on the tab:

SLO details

  • Compliance (time window duration).

  • Target SLO

  • Current SLI

  • Error Budget Remaining

Service Level Indicator

The Service Level Indicator metric chart incorporates SLIs that are essential for measuring the performance and reliability of your services. It represents a measurable metric that reflects the health or quality of a specific aspect of your service, it can also act as a proxy for the user experience and helps you set reliability targets.

Based on the entity type, the metric is chosen and the calculation is done.

  • Threshold metrics (raw metrics): Threshold metrics operate based on a single time series. These metrics focus on a specific value (for example, response time, error rate) and compare it against a predefined threshold:

    • Latency Threshold: Ensure that 95% of requests are processed within 200 milliseconds.

    • Error Rate Threshold: Keep the error rate below 1% for successful responses.

  • Ratio metrics (count metrics): Ratio metrics operate based on two-time series, a count of good events (successful outcomes) and a count of total events (both successful and unsuccessful). These metrics provide a more nuanced view by considering both success and failure events. The ratio is calculated using the following formula:

    SLI = (good events / total events) * 100.

    • Success Rate Ratio: Measure the percentage of successful requests out of the total requests.

Burn Rate

The burn rate is a measure that calculates how fast you use up your error budget. A burn rate above 1 indicates that the error budget is being consumed faster than expected.

Burn Down

This chart displays the error budget consumption based on the internal query that calculates the error budget consumed over a specific time window (for example, 30 days).