# How Infrastructure Performance for AWS Network Manager works How Infrastructure Performance works Network latency measurements are an aggregate of latency performance information captured by active probing between AWS managed probes within the AWS global network. Using Infrastructure Performance,AWS Network Manager allows you to view network performance based on the following metrics: + Inter-Region latency metrics, generated by aggregating latency measurements from probes located across AWS Regions, filtered on your chosen source and destination AWS Region pairs. + Inter-Availability Zone latency metrics, generated by aggregating latency measurements from probes located across Availability Zones, filtered on your chosen source and destination Availability Zones. + Intra-Availability Zone latency metrics, generated by aggregating latency measurements between all probes within a single Availability Zone. This includes all probes that are across AWS data centers, and within the same data center, for the chosen Availability Zone. The placement of the probes is not related to EC2 instances or AWS services running in your account. For these reasons, the same latency metric data is presented in all AWS accounts. ## How latency is calculated Metric data is generated by computing the median (`P50`) of all latency measurements from AWS managed probes for every five-minute (`5M`) interval. Latency metrics represent round-trip traffic latency of synthetic traffic streams between selected sources and destinations. This is a combination of the latency of the underlying physical network and the latency between the host and the underlying physical network. The Infrastructure Performance dashboard also provides a network health status, indicating whether the aggregate network latency is normal or degraded. The health status is determined based on expected AWS network performance between AWS Regions and Availability Zones. The health status is available only for inter-Region and inter-Availability Zone metrics. **Note** In case of a degraded health status, AWS is automatically notified of the incident. ## Region availability Infrastructure Performance is available in the following AWS Regions. If a Region is not listed here it is not available: | AWS Region | Description | | --- | --- | | us-east-1 | US East (N. Virginia) | | us-east-2 | US East (Ohio) | | us-west-1 | US West (N. California) | | us-west-2 | US West (Oregon) | | ap-east-1 | Asia Pacific (Hong Kong) | | ap-south-1 | Asia Pacific (Mumbai) | | ap-northeast-3 | Asia Pacific (Osaka) | | ap-northeast-2 | Asia Pacific (Seoul) | | ap-southeast-1 | Asia Pacific (Singapore) | | ap-southeast-2 | Asia Pacific (Sydney) | | ap-northeast-1 | Asia Pacific (Tokyo) | | ap-southeast-3 | Asia Pacific (Jakarta) | | ap-southeast-4 | Asia Pacific (Melbourne) | | ca-central-1 | Canada (Central) | | eu-central-1 | Europe (Frankfurt) | | eu-cenral-2 | Europe (Zurich) | | eu-west-1 | Europe (Ireland) | | eu-west-2 | Europe (London) | | eu-west-3 | Europe (Paris) | | eu-north-1 | Europe (Stockholm) | | eu-south-1 | Europe (Milan) | | eu-south-2 | Europe (Spain) | | sa-east-1 | South America (São Paulo) | | af-south-1 | Africa (Cape Town) | | me-south-1 | Middle East (Bahrain) | # Monitor Infrastructure Performance using the AWS Network Manager console Monitor network performance using the console View the network performance between two Regions, between two Availability Zones, or in the two connections within an Availability Zone. After you choose a starting and ending date, and a period by which to aggregate performance, the network performance metrics display in the **Health status timeline** and **Network latency timeline** for that time period. You can also subscribe or unsubscribe performance metrics from Amazon CloudWatch. When you subscribe, you're able to view performance metrics in CloudWatch. **Note** Enabling subscriptions with CloudWatch might incur a charge. For more information about CloudWatch pricing guidelines, see [Amazon CloudWatch pricing](https://aws.amazon.com/cloudwatch/pricing/). Choose between two Regions, two Availability Zones, or connections within an Availability Zone. You can choose multiple Region, Availability Zones or intra-Availability Zones, but you can't have a combination of all three. Combinations of Regions, Inter-Availability Zones, or Intra-Availability Zones are not supported. **To choose a Region or Availability Zone** 1. Access the Network Manager console at [https://console.aws.amazon.com/networkmanager/home/](https://console.aws.amazon.com/networkmanager/home). 1. Under **Monitoring and troubleshooting**, choose **Infrastructure Performance**. 1. From the Regions drop-down list, choose whether you want to see **Inter-Region**, **Inter-Availability Zone**, or **Intra-Availability Zone** performance data. By default the table populates all possible **Source** and **Destination** pairs of Regions or Availability Zones, or all possible pairs within an intra-Availability Zone. This section displays whether that pair is subscribed to Amazon CloudWatch. Sort on **Source**, **Destination**, or **Subscription** to more easily select which Regions or Availability Zones you want to check network performance for. 1. Scroll through the pages and choose one or more pairs. 1. (Optional). To choose specific Regions or Availability Zone, choose **Find metrics**. 1. In the **Find metrics** field, choose one of the following **Properties**: + **Source** — The Region or Inter/Intra Availability Zone that's the source of the network traffic. + **Destination** — The Region or Inter/Intra Availability Zone that's the destination for the network source. + **Subscription** — Indicates whether the Region or Availability Zone is **Subscribed** or **Unsubscribed** for CloudWatch notifications. For more information about managing subscriptions with CloudWatch, see [Amazon CloudWatch subscriptions for Infrastructure Performance](nmip-subscriptions-cw.md). 1. For the property operator, choose one of the following operators: + **Source =** — Returns network performance results only for the chosen Region, Availability Zone, or an **unsubscribed** subscription status. + **Source \$1=** — Does not equal. Excludes the chosen Region, Availability Zone, or **unsubscribed** subscription status from the network performance results. 1. From the **Source Values** drop-down list, choose the value to apply to the operator. 1. Repeat the previous step if you want to include multiple metrics in the results. 1. By default, each metric that you add is separated by **and** connecting logic, restricting results to the criteria that meet all of the requirements. To return results that match any of your chosen metrics, choose **or** from the drop-down list. All metrics use the same connecting logic. Combinations of **and**/**or** logic are not supported. If you change the connecting logic between two metrics, all other metrics will use the same connecting logic. 1. The following example shows a filter for a single Source Region, `us-east-1` ***and*** a single Destination Region, `eu-central-1`. The results return one combination: ![\[Infrastructure Performance filter for a single source and a single destination .\]](http://docs.aws.amazon.com/network-manager/latest/infrastructure-performance/images/nmip-metric-filter.png) 1. Choose a time range, and then choose **Apply**. You can choose a **Relative range** or an **Absolute range**. **Relative range** options use a date range based on your current date and time: + **Last 30 minutes** + **Last 1 hour** + **Last 6 hour** + **Last 1 day** + **Last 1 week** + **Custom range** — Set a custom date range. Enter a **Duration**, and then choose one of the following **Units of time**: + **seconds** + **minutes** (default) + **hours** + **days** + **weeks** + **years** **Absolute range **allows you to set a specific start date and time and end date and time. The date range cannot exceed 30 days. **Note** Performance metrics are not available prior to Oct 26, 2022 00:00:00 GMT\$1000. + On the calendar control, choose the start and end dates. Or, + Enter a **Start date** and **End date** using a **YYYY/MM/DD** format. + Enter a **Start time** and **End time** using an **hh:mm** format. If you choose dates using the calendar control, the default start and end times are set to **00:00**. You can change those times by entering your preferred times in the fields. When using an absolute range, both start and end dates are required. 1. Choose the time period by which you want the performance aggregated. Options are: + **5M** — Five minutes (default) + **15M** — Fifteen minutes + **1H** — One hour + **1D** — One day + **1W** — One week 1. The **Health status timeline** and **Network latency sections** update. ## Health status timeline The **Health status timeline** shows a single consolidated view of network performance for the chosen Region, Inter-Availability Zone, or Intra-Availability Zone pairs over the starting and end times, consolidated by the aggregation period. The timeline displays the following network latency statuses: + **Healthy** (green) — Network performance fell within the expected range. + **Degraded** (yellow) — Network latency fell below the expected range. Hover over the status bar at a particular date and time to display the health status at that time for the pairs you've chosen. In the following example, network performance is checked for two pairs of Regions: `eu-north-1` and `me-south-1`, as well as `eu-north-1` and `ap-south-1`. The time period uses a **Relative time** of **Last 1 hour** on November 11, and the aggregation period is **5M** (five minutes). From **11:55** AM to 12:05 PM during this hour there was **Degraded** performance. Hovering over the status bar shows that degraded performance occurred only between the `eu-north-1` and `me-south-1` Regions, while network performance for the other Region pair was healthy. **Note** This example is provided for illustrative purposes only and does not represent the actual network health status between these Regions during this time. ![\[The Health status timeline for two pairs of Regions, one pair showing degraded latency.\]](http://docs.aws.amazon.com/network-manager/latest/infrastructure-performance/images/nmip-degraded.png) ## Network latency The Network latency section shows the actual network performance speed between the chosen Region, Availability Zone, or within an Availability Zone pairs over the starting and end times, consolidated by the aggregation period. Latency can run from a minimum of **0 ms** to a maximum of **160 ms**. Each chosen pair is represented by individual lines in the section, with a legend at the bottom of the page mapping a line to a pair. Hover anywhere over the graph at a particular time to view a comprehensive list of all chosen pairs and the actual latency at that time. Or, if you want to see the latency for a specific pair, choose that line. The following example shows the latency of two Region pairs : `eu-north-1` and `me-south-1`, as well as `eu-north-1` and `ap-south-1` at **02:15** on **Nov 11**. **Note** This example is provided for illustrative purposes only and does not represent the actual network latency between these Regions during this time. ![\[Network latency status, showing the actual network performance for two sets of Region pairs.\]](http://docs.aws.amazon.com/network-manager/latest/infrastructure-performance/images/nmip-latency-all.png) # Monitor Infrastructure Performance using the AWS CLI Monitor network performance using the CLI [GetAwsNetworkPerformanceData](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_GetAwsNetworkPerformanceData.html) returns network performance metrics. For more information, see [GetAwsNetworkPerformanceData](https://docs.aws.amazon.com/AWSEC2/latest/APIReference/API_GetAwsNetworkPerformanceData.html). In the following example, network performance is queried between two Regions, `us-east-1` and `us-west-2` on `2022-10-26` between a start-time of `12:00:00` and an end time of `12:30:00`. The request also uses the following parameters: + `Metric` indicates what type of metric is being requested. In this example, `aggregate-latency`, indicates that the performance metrics are aggregated and returned for latency. + `Statistic` is the median value of the metric. In this example, `p50` is the statistic of all the data points gathered within those five minutes. **Note** `p50` is the only supported statistic. + `Period` indicates the interval at which performance the `metric` is returned. In this example, `aggregated-latency` metrics are returned for every `five-minutes`. ``` aws ec2 --region ap-southeast-3 get-aws-network-performance-data --start-time 2022-10-26T12:00:00.000Z --end-time 2022-10-26T12:30:00.000Z --data-queries Id=id1,Source=us-east-1,Destination=us-west-2,Metric=aggregate-latency,Statistic=p50,Period=five-minutes ``` The results return the following: ``` { "DataResponses": [ { "Id": "id1", "Source": "us-east-1", "Destination": "us-west-2", "Metric": "aggregate-latency", "Statistic": "p50", "Period": "five-minutes", "MetricPoints": [ { "StartDate": "2022-10-26T12:00:00+00:00", "EndDate": "2022-10-26T12:05:00+00:00", "Value": 62.44349, "Status": "OK" }, { "StartDate": "2022-10-26T12:05:00+00:00", "EndDate": "2022-10-26T12:10:00+00:00", "Value": 62.483498, "Status": "OK" }, { "StartDate": "2022-10-26T12:10:00+00:00", "EndDate": "2022-10-26T12:15:00+00:00", "Value": 62.51248, "Status": "OK" }, { "StartDate": "2022-10-26T12:15:00+00:00", "EndDate": "2022-10-26T12:20:00+00:00", "Value": 62.635475, "Status": "OK" }, { "StartDate": "2022-10-26T12:20:00+00:00", "EndDate": "2022-10-26T12:25:00+00:00", "Value": 62.733974, "Status": "OK" }, { "StartDate": "2022-10-26T12:25:00+00:00", "EndDate": "2022-10-26T12:30:00+00:00", "Value": 62.773975, "Status": "OK" }, { "StartDate": "2022-10-26T12:30:00+00:00", "EndDate": "2022-10-26T12:35:00+00:00", "Value": 62.75349, "Status": "OK" } ] } ] } ```