Skip to content

Latest commit

 

History

History
97 lines (62 loc) · 5.22 KB

api-management-capacity.md

File metadata and controls

97 lines (62 loc) · 5.22 KB
title description services documentationcenter author manager editor ms.service ms.workload ms.topic ms.date ms.author
Capacity of an Azure API Management instance | Microsoft Docs
This article explains what the capacity metric is and how to make informed decisions whether to scale an Azure API Management instance.
api-management
mikebudzynski
anneta
api-management
integration
article
06/18/2018
apimpm

Capacity of an Azure API Management instance

Capacity is the single, most important Azure Monitor metric for making informed decisions whether to scale an API Management instance to accommodate more load. Its construction is complex and imposes certain behavior.

This article explains what the capacity is and how it behaves. It shows how to access capacity metrics in the Azure portal and suggests when to consider scaling or upgrading your API Management instance.

Prerequisites

To follow the steps from this article, you must have:

[!INCLUDE premium-dev-standard-basic.md]

What is capacity

Capacity metric

Capacity is an indicator of load on an APIM instance. It reflects resources usage (CPU, memory) and network queue lengths. CPU and memory usage reveals resources consumption by:

  • APIM services, such as management actions or request processing, which can include forwarding requests or running a policy
  • selected operating system processes, including processes that involve cost of SSL handshakes on new connections.

Total capacity is an average of its own values from every unit of an API Management instance.

Capacity metric behavior

Because of its construction, in real life capacity can be impacted by many variables, for example:

  • connection patterns (new connection on a request vs reusing the existing connection)
  • size of a request and response
  • policies configured on each API or number of clients sending requests.

The more complex operations on the requests are, the higher the capacity consumption will be. For example, complex transformation policies consume much more CPU than a simple request forwarding. Slow backend service responses will increase it too.

Important

Capacity is not a direct measure of the number of requests being processed.

Capacity metric spikes

Capacity can also spike intermittently or be greater than zero even if there are no requests being processed. It happens because of system- or platform-specific actions and should not be taken into consideration when deciding whether to scale an instance.

Use the Azure Portal to examine capacity

Capacity metric

  1. Navigate to your APIM instance in the Azure portal.

  2. Select Metrics (preview).

  3. From the purple section, select Capacity metric from available metrics and leave the default Avg aggregation.

    [!TIP] You should always look at a capacity metric breakdown per location to avoid wrong interpretations.

  4. From the green section, select Location for splitting the metric by dimension.

  5. Pick a desired timeframe from the top bar of the section.

    You can set a metric alert to let you know when something unexpected is happening. For example, get notifications when your APIM instance has been exceededing its expected peak capacity for over 20 minutes.

    [!TIP] You can configure alerts to let you know when your service is running low on capacity or use Azure Monitor autoscaling functionality to automatically add an Azure API Management unit. Scaling operation can take around 30 minutes, so you should plan your rules accordingly.
    Only scaling the master location is allowed.

Use capacity for scaling decisions

Capacity is the metric for making decisions whether to scale an API Management instance to accommodate more load. Consider:

  • Looking at a long-term trend and average.
  • Ignoring sudden spikes that are most likely not related to any increase in load (see "Capacity metric behavior" section for explanation).
  • Upgrading or scaling your instance, when capacity's value exceeds 60% or 70% for a longer period of time (for example 30 minutes). Different values may work better for your service or scenario.

Tip

If you are able to estimate your traffic beforehand, test your APIM instance on workloads you expect. You can increase the request load on your tenant gradually and monitor what value of the capacity metric corresponds to your peak load. Follow the steps from the previous section to use Azure portal to understand how much capacity is used at any given time.

Next steps

How to scale or upgrade an Azure API Management service instance