New Time Slice SLO (#20888)

* Add time slice to left nav * Add time slice instructions and images * Add uptime calculations page * Add uptime calculations to left nav * Standardize use of Time Slice SLO * Remove duplicate file * Merge uptime with time slice * Add SLO comparison chart * Apply code review suggestions * Update content/en/service_management/service_level_objectives/_index.md * Apply suggestions from code review Co-authored-by: jhgilbert <[email protected]> * Apply suggestions from code review, removed commented examples * Add time slice to left nav * Add time slice instructions and images * Add uptime calculations page * Add uptime calculations to left nav * Standardize use of Time Slice SLO * Remove duplicate file * Merge uptime with time slice * Add SLO comparison chart * Apply code review suggestions * Update content/en/service_management/service_level_objectives/_index.md * Apply suggestions from code review Co-authored-by: jhgilbert <[email protected]> * Apply suggestions from code review, removed commented examples * Add examples with images * minor changes * API info comparison chart * update comparison chart * update comparison chart again * fix status correction info * update SLO definitions * calendar view info --------- Co-authored-by: jhgilbert <[email protected]> Co-authored-by: Roxanne Moslehi <[email protected]>
DataDog · Jan 17, 2024 · aa94230 · aa94230
1 parent 4b0e778
commit aa94230
Show file tree

Hide file tree

Showing 12 changed files with 218 additions and 46 deletions.
diff --git a/config/_default/menus/menus.en.yaml b/config/_default/menus/menus.en.yaml
@@ -1333,16 +1333,21 @@ main:
     parent: slos
     identifier: slos_metric
     weight: 2
+  - name: Time Slice SLOs
+    url: service_management/service_level_objectives/time_slice/
+    parent: slos
+    identifier: slos_time_slice
+    weight: 3
   - name: Error Budget Alerts
     url: service_management/service_level_objectives/error_budget/
     parent: slos
     identifier: error_budget
-    weight: 3
+    weight: 4
   - name: Burn Rate Alerts
     url: service_management/service_level_objectives/burn_rate/
     parent: slos
     identifier: burn_rate
-    weight: 4
+    weight: 5
   - name: Guides
     url: service_management/service_level_objectives/guide/
     parent: slos

diff --git a/content/en/service_management/service_level_objectives/_index.md b/content/en/service_management/service_level_objectives/_index.md
diff --git a/content/en/service_management/service_level_objectives/guide/_index.md b/content/en/service_management/service_level_objectives/guide/_index.md
@@ -7,6 +7,7 @@ disable_toc: true
 
 {{< whatsnext desc="General guides:">}}
     {{< nextlink href="/service_management/service_level_objectives/guide/slo-checklist" >}}SLO Checklist{{< /nextlink >}}
+    {{< nextlink href="/service_management/service_level_objectives/guide/slo_types_comparison" >}}SLO Type Comparison{{< /nextlink >}}
 {{< /whatsnext >}}
 
 {{< whatsnext desc="Dashboard guides:">}}

diff --git a/...nt/en/service_management/service_level_objectives/guide/slo_types_comparison.md b/...nt/en/service_management/service_level_objectives/guide/slo_types_comparison.md
@@ -0,0 +1,53 @@
+---
+title: SLO Type Comparison
+kind: Guide
+further_reading:
+- link: "/service_management/service_level_objectives/"
+  tag: "Documentation"
+  text: "Overview of Service Level Objectives"
+- link: "/service_management/service_level_objectives/metric/"
+  tag: "Documentation"
+  text: "Metric-based SLOs"
+- link: "/service_management/service_level_objectives/monitor/"
+  tag: "Documentation"
+  text: "Monitor-based SLOs"
+- link: "/service_management/service_level_objectives/time_slice/"
+  tag: "Documentation"
+  text: "Time Slice SLOs"
+---
+
+## Overview
+
+When creating SLOs, you can choose from the following types:
+- **Metric-based SLOs**: can be used when you want the SLI calculation to be count-based, the SLI is calculated as the sum of good events divided by the sum of total events.
+- **Monitor-based SLOs**: can be used when you want the SLI calculation to be time-based, the SLI is based on the Monitor's uptime. Monitor-based SLOs must be based on a new or existing Datadog monitor, any adjustments must be made to the underlying monitor (cannot be done through SLO creation).
+- **Time Slice SLOs**: can be used when you want the SLI calculation to be time-based, the SLI is based on your custom uptime definition (amount of time your system exhibits good behavior divided by the total time). Time Slice SLOs do not require a Datadog monitor, you can try out different metric filters and thresholds and instantly explore downtime during SLO creation.
+
+## Comparison chart
+
+|                                                                       | **Metric-based SLO**                                                                                                                      | **Monitor-based SLO**                                                                                                                                               | **Time Slice SLO**                                                                   |
+|-----------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------------|
+| **Supported data types**                                              | Metrics with type of count, rate, or distribution                                                                                         | Metric Monitor types, Synthetic Monitors, and Service Checks                                                                                                        | All metric types (including gauge metrics)                                           |
+| **Functionality for SLO with Groups**                                               | SLO calculated based on all groups<br><br>Can view all groups in SLO side panel, up to 20 groups in SLO summary widget                                                                                                                  | Supported for SLOs with a single multi alert Monitor<br><br>**Option 1:** SLO calculated based on all groups (can view 5 worst groups in SLO side panel and SLO summary widget)<br>**Option 2:** SLO calculated based on up to 20 selected groups (can view all selected groups in SLO side panel and SLO summary widget)                                                                                                                                | SLO calculated based on all groups<br><br>Can view all groups in SLO side panel, up to 20 groups in SLO summary widget                                                           |
+| **SLO details side panel (up to 90 days of historical data)** | Can set custom time windows to view SLO info                                                                                              | Cannot set custom time windows to view SLO info (can view 7, 30, or 90 day history)                                                                                                                     | Can set custom time windows to view SLO info                                         |
+| **SLO alerting ([Error Budget][1] or [Burn Rate][2] Alerts)**         | Available                                                                                                      | Available for SLOs based on Metric Monitor types only (not available for Synthetic Monitors or Service Checks)                                                      | Not available                                                    |
+| [**SLO Status Corrections**][3]                                       | Correction periods are ignored from SLO status calculation                                                              | Correction periods are ignored from SLO status calculation                                                                                        | Correction periods are counted as uptime in SLO status calculation |
+| **[SLO Widgets][4] (up to 90 days of historical data)**                                                  | Available                                                                                                           | Available                                                                                                                           | Available                                                        |
+| [**SLO Data Source**][5]                                              | Available (with up to 15 months of historical data)                                                                | Not available                                                                                                                                     | Not available                                                |
+| **Handling missing data in the SLO calculation**                      | Missing data is ignored in SLO status and error budget calculations                                                                       | Missing data is handled based on the [underlying Monitor's configuration][6]                                                                                        | Missing data is treated as uptime in SLO status and error budget calculations        |
+| **Uptime Calculations**                                          |  N/A                                                                                  |  Uptime calculations are based on the underlying Monitor <br><br>If groups are present, overall uptime requires *all* groups to have uptime| [Uptime][7] is calculated by looking at discrete time chunks, not rolling time windows<br><br>If groups are present, overall uptime requires *all* groups to have uptime |
+| **Calendar View on SLO Manage Page**                                   | Available                                                                                                                                | Not available                                                                                                                                                      | Available                                                                            |
+| **Public [APIs][8] and Terraform Support**                                   | Available                                                                                                                                 | Available                                                                                                                                                     | Not available                                                                            |
+
+## Further Reading
+
+{{< partial name="whats-next/whats-next.html" >}}
+
+[1]: https://docs.datadoghq.com/service_management/service_level_objectives/error_budget/
+[2]: https://docs.datadoghq.com/service_management/service_level_objectives/burn_rate/
+[3]: https://docs.datadoghq.com/service_management/service_level_objectives/#slo-status-corrections
+[4]: https://docs.datadoghq.com/service_management/service_level_objectives/#slo-widgets
+[5]: https://docs.datadoghq.com/dashboards/guide/slo_data_source/
+[6]: https://docs.datadoghq.com/service_management/service_level_objectives/monitor/#missing-data
+[7]: /service_management/service_level_objectives/time_slice/#uptime-calculations
+[8]: https://docs.datadoghq.com/api/latest/service-level-objectives/
diff --git a/content/en/service_management/service_level_objectives/time_slice.md b/content/en/service_management/service_level_objectives/time_slice.md
@@ -0,0 +1,95 @@
+---
+title: Time Slice SLOs
+kind: documentation
+is_beta: true
+further_reading:
+- link: "service_management/service_level_objectives/"
+  tag: "Documentation"
+  text: "Overview of Service Level Objectives"
+---
+
+{{< jqmath-vanilla >}}
+
+## Overview
+
+Time Slice SLOs allow you to measure reliability using a custom definition of uptime. You define uptime as a condition over a metric timeseries. For example, you can create a latency SLO by defining uptime as whenever p95 latency is less than 1 second.
+
+Time Slice SLOs are a convenient alternative to Monitor-based SLOs. You can create an uptime SLO without going through a monitor, so you don't have to create and maintain both a monitor and an SLO. 
+
+## Create a Time Slice SLO
+
+You can create a Time Slice SLO through the following ways:
+- [Create an SLO from the create page](#create-an-slo-from-the-create-page)
+- [Export an existing Monitor-based SLO](#export-an-existing-monitor-slo)
+- [Import from a monitor](#import-from-a-monitor)
+
+### Create an SLO from the create page
+
+{{< img src="service_management/service_level_objectives/time_slice/create_and_configuration.png" alt="Configuration options to create a Time Slice SLO" style="width:100%;" >}}
+
+1. Navigate to [**Service Management > SLOs**][1].
+1. Click **+ New SLO** to open up the Create SLO page.
+1. Select **By Time Slices** to define your SLo measurement. 
+1. Define your uptime condition by choosing a metric query, comparator and threshold. For example, to define uptime as whenever p95 latency is less than 1s. Alternatively, you can [import the uptime from a monitor](#import-from-a-monitor).
+1. Choose your timeframe and target
+1. Name and tag your SLO.
+1. Click **Create**.
+
+### Export an existing monitor SLO
+
+<div class="alert alert-info">Only single metric monitor SLOs can be exported. Non-metric monitors or multi-monitor SLOs cannot be exported.</div>
+
+Create a Time Slice SLO by exporting an existing Monitor-based SLO. From a monitor SLO, click **Export to Time Slice SLO**.
+
+{{< img src="service_management/service_level_objectives/time_slice/export_monitor_slo.png" alt="On a Monitor-based SLO detail side panel, the button to Export to Time Slice is highlighted" style="width:90%;" >}}
+
+### Import from a monitor
+
+<div class="alert alert-info">Only metric monitor SLOs appear in the monitor selection for import. </div>
+
+From the **Create or Edit SLO** page, under **Define your SLI**, click **Import from Monitor** and select from the dropdown or search in the monitor selector.
+
+**Note**: Time Slice SLOs do not support rolling periods. Rolling periods do not transfer from a monitor query to a Time Slice query. 
+
+{{< img src="service_management/service_level_objectives/time_slice/import_from_monitor.png" alt="Highlighted option to Import From Monitor in the Define your SLI section of an SLO configuration" style="width:90%;" >}}
+
+## Uptime calculations
+
+To calculate the uptime percentage for a Time Slice SLOs, Datadog cuts the timeseries into equal-duration intervals, called "slices". The length of the interval is 5 minutes and not configurable. The space and time aggregation are determined by the metric query. For more information on time and space aggregation, see the [metrics][2] documentation. 
+
+For each slice, there is a single value for the timeseries, and the uptime condition (such as `value < 1`) is evaluated for each slice. If the condition is met, the slice is considered uptime, otherwise it is considered downtime.
+
+{{< img src="service_management/service_level_objectives/time_slice/uptime_latency.png" alt="Time Slice SLO detail panel showing application latency with one uptime violation" style="width:100%;" >}}
+
+For the above example, exactly one point in the timeseries violates the uptime condition (in this case, the condition is that the p95 latency is less than or equal to 2.5 seconds). Since the total time period shown is 12 hours (720 minutes), and 715 minutes are considered uptime (720 min total time - 5 min downtime), the uptime percentage is 715/720 * 100 = 99.305%
+
+### Groups and overall uptime
+
+Time Slice SLOs allow you to track uptime for individual groups, where groups are defined in the "group by" portion of the metric query. 
+
+When groups are present, uptime is calculated for each individual group. However, overall uptime works differently. In order to match existing monitor SLO functionality, Time Slice SLOs use the same definition of overall uptime. When **all** groups have uptime, it is considered overall uptime. Conversely, if **any** group has downtime, it is considered overall downtime. Overall uptime is always less than the uptime for any individual group.
+
+{{< img src="service_management/service_level_objectives/time_slice/uptime_latency_groups.png" alt="Time Slice SLO detail panel of application latency uptime with groups" style="width:100%;" >}}
+
+In the example above, environment "prod" has 5 minutes of downtime over a 12 hour (720 minute) period, resulting in approximately 715/720 * 100 = 99.305% of uptime. Environment "dev" also had 5 minutes of downtime, resulting in the same uptime. This means that overall downtime--when either datacenter prod or dev had downtime--was 10 minutes (since there is no overlap), resulting in approximately (720-10)/720 * 100 = 98.611% uptime.
+
+### Corrections
+
+Time Slice SLOs count correction periods as uptime in all calculations. Since the total time remains constant, the error budget is always a fixed amount of time as well. This is a significant simplification and improvement over how corrections are handled for monitor-based SLOs.
+
+For monitor-based SLOs, corrections are periods that are removed from the calculation. If a one-day-long correction is added to a 7-day SLO, 1 hour of downtime counts as 0.7% instead of 0.6%:
+
+$$ 60/8640 *100 = ~0.7% $$
+
+The effects on error budget can be unusual. Removing time from an uptime SLO causes time dilation, where each minute of downtime represents a larger fraction of the total time. 
+
+### Missing data
+
+In Time Slice SLOs, missing data is always treated as uptime. While missing data is treated as uptime, it is gray on the timeline visualization.
+
+## Further Reading
+
+{{< partial name="whats-next/whats-next.html" >}}
+
+[1]: https://app.datadoghq.com/slo/manage
+[2]: /metrics/#time-and-space-aggregation
diff --git a/static/images/service_management/service_level_objectives/calendar-view-slo.png b/static/images/service_management/service_level_objectives/calendar-view-slo.png
diff --git a/...ice_management/service_level_objectives/time_slice/create_and_configuration.png b/...ice_management/service_level_objectives/time_slice/create_and_configuration.png
diff --git a/...s/service_management/service_level_objectives/time_slice/export_monitor_slo.png b/...s/service_management/service_level_objectives/time_slice/export_monitor_slo.png
diff --git a/.../service_management/service_level_objectives/time_slice/import_from_monitor.png b/.../service_management/service_level_objectives/time_slice/import_from_monitor.png
diff --git a/...anagement/service_level_objectives/time_slice/time_slice_detail_panel_group.png b/...anagement/service_level_objectives/time_slice/time_slice_detail_panel_group.png
diff --git a/...mages/service_management/service_level_objectives/time_slice/uptime_latency.png b/...mages/service_management/service_level_objectives/time_slice/uptime_latency.png
diff --git a/...ervice_management/service_level_objectives/time_slice/uptime_latency_groups.png b/...ervice_management/service_level_objectives/time_slice/uptime_latency_groups.png