Stay organized with collectionsSave and categorize content based on your preferences.
Creating SLOs
This page describes how to create a service level objective (SLO) in the
Google Cloud console. For information on creating a SLO programmatically,
seeWorking with the SLO API
For an overview and information on designing SLOs, see the following:
Select the Google Cloud project from the drop-down list on the menu bar.
Click the service that you want to create an SLO for.
In the left navigation bar, clickHealth.
Click theCreate an SLOlink.
ClickSet your service-level indicator (SLI)to select the type of service level indicator (SLI)
to track for this SLO. Choose one of the following:
Availability: The ratio of the number of successful
responses to the number of all responses. Requests that fail before they
reach the Envoy sidecar proxy for your service (because of networking or
DNS failures, for example) aren't included in this ratio.
Latency: The ratio of the number of calls that are below the
specifiedLatency Thresholdto the number of all calls.
For latency SLIs, enter theLatency Thresholdin milliseconds.
In thePerformance Goalsection, enter a percentage in theGoalfield to set the performance target for the SLI. Cloud Service Mesh
uses this value to calculate theerror budgetyou have for this SLO.
In theCompliance Periodsection, select thePeriod Typeand thePeriod Length. SeeCompliance periodsfor more information on these settings.
Alternatively, in theSet your SLIsection, you may selectWindows-basedSLI. A windowed SLI can help you catch
periods of time when the service won't meet the SLOCompliance target(such as when there are spikes in the number of requests that increase
latency for a short period of time). When you select this option, you must
specify:
Goodness criterion: Set a threshold for the percentage of "good performance" that must be met in order for a given window to count as "good."
Duration: Set the length of each window, over which SLO
performance will be measured in increments during the compliance period.
For example, suppose you have anAvailabilitySLO with aRolling7day period, and anSLO Goalof99%. Then you add
aGoodness criterionof95% and aDurationof5minutes.
To be compliant, the service needs95% of all5minute windows
over the last7days to be available at least99% of the time.
Optionally, clickName your SLOto change the default SLO display
name. Cloud Service Mesh supplies a default name that describes the SLO
based on the settings.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-09-04 UTC."],[],[],null,["# Creating SLOs\n=============\n\nThis page describes how to create a service level objective (SLO) in the\nGoogle Cloud console. For information on creating a SLO programmatically,\nsee\n[Working with the SLO API](/stackdriver/docs/solutions/slo-monitoring/api/api-structures)\n\nFor an overview and information on designing SLOs, see the following:\n\n- [Service level objectives overview](/service-mesh/v1.20/docs/observability/slo-overview)\n- [Designing SLOs](/service-mesh/v1.20/docs/observability/design-slo)\n\nTo create an SLO:\n\n1. Go to the **Health** tab for a service:\n\n 1. In the Google Cloud console, go to **Cloud Service Mesh**.\n\n [Go to Cloud Service Mesh](https://console.cloud.google.com/anthos/services)\n 2. Select the Google Cloud project from the drop-down list on the menu bar.\n\n 3. Click the service that you want to create an SLO for.\n\n 4. In the left navigation bar, click **Health**.\n\n2. Click the **Create an SLO** link.\n\n3. Click **Set your service-level indicator (SLI)** to select the type of service level indicator (SLI)\n to track for this SLO. Choose one of the following:\n\n - **Availability**: The ratio of the number of successful responses to the number of all responses. Requests that fail before they reach the Envoy sidecar proxy for your service (because of networking or DNS failures, for example) aren't included in this ratio.\n - **Latency** : The ratio of the number of calls that are below the specified **Latency Threshold** to the number of all calls.\n4. For latency SLIs, enter the **Latency Threshold** in milliseconds.\n\n5. In the **Performance Goal** section, enter a percentage in the **Goal**\n field to set the performance target for the SLI. Cloud Service Mesh\n uses this value to calculate the\n [error budget](/service-mesh/v1.20/docs/observability/design-slo#error_budgets)\n you have for this SLO.\n\n6. In the **Compliance Period** section, select the **Period Type** and the\n **Period Length** . See\n [Compliance periods](/service-mesh/v1.20/docs/observability/design-slo#compliance_periods)\n for more information on these settings.\n\n7. Alternatively, in the **Set your SLI** section, you may select **Windows-based** SLI. A windowed SLI can help you catch\n periods of time when the service won't meet the SLO **Compliance target**\n (such as when there are spikes in the number of requests that increase\n latency for a short period of time). When you select this option, you must\n specify:\n\n - **Goodness criterion**: Set a threshold for the percentage of \"good performance\" that must be met in order for a given window to count as \"good.\"\n - **Duration**: Set the length of each window, over which SLO performance will be measured in increments during the compliance period.\n\n For example, suppose you have an **Availability** SLO with a\n **Rolling** **7** day period, and an **SLO Goal** of **99** %. Then you add\n a **Goodness criterion** of **95** % and a **Duration** of **5** minutes.\n To be compliant, the service needs **95** % of all **5** minute windows\n over the last **7** days to be available at least **99**% of the time.\n8. Optionally, click **Name your SLO** to change the default SLO display\n name. Cloud Service Mesh supplies a default name that describes the SLO\n based on the settings.\n\n9. Click **Submit**.\n\nWhat's next\n-----------\n\n- [Monitoring SLOs](/service-mesh/v1.20/docs/observability/monitor-slo)\n- [Creating an alerting policy on an SLO](/service-mesh/v1.20/docs/observability/alert-policy-slo)"]]