edit-icon download-icon

Infrastructure monitoring

Last Updated: Jan 16, 2018

EDAS collects data from the ECS instances that run applications and provides the single-instance and cluster views of the CPU, memory, load, network, and disk metrics based on the analysis results. Data in all monitoring views is collected and processed from the application point of view.

Note: A latency exists from data collection to analysis. Therefore, EDAS cannot provide exactly real-time monitoring views. The current latency is about two minutes.

Perform the following steps to view cluster or single-instance statistics:

  1. Log on to the EDAS console, click Applications, and click an application in the application list.

  2. On the application details page, select Application Monitoring > Infrastructure Monitoring on the left-side menu bar to go to the basic monitoring page.

    On the basic monitoring page, group data in the latest half an hour is monitored by default.

    You can select a time range to monitor group data for a different time range, or select the Single Instance Data tab to see single-instance data.

  3. Select a type of monitored data.

    Data to be monitored includes group data and single-instance data.

    The following metrics of the two data types are monitored from different dimensions:

    • CPU data: Indicates the CPU usage, which is the sum of the user usage and system usage. The group data graph displays the average value of the usage of all ECS instances in the application group.

    • Memory data: Indicates the total size and actual usage of the physical memory. The group data graph displays the total size and total usage of all ECS instances in the application group.

    • Load data: Indicates the “1 min load” field in the system load. The group data graph displays the average value of “1 min load” of all ECS instances in the application group.

    • Network speed data: Indicates the write/read speed of the network card. If an ECS instance contains multiple network cards, the data indicates the sum of write/read speeds of all network cards whose names start with “eth”. The group data graph displays the average value of all ECS instances in the application group.

    • Disk data: Indicates the total size and actual usage of all disks attached to the ECS instance. The group data graph displays the total size and total usage of all ECS instances in the application group.

    • Disk read/write speed: Indicates the sum of the read/write speeds of all disks attached to the ECS instance. The group data graph displays the average value of the data of all ECS instances in the application group.

    • Disk read/write numbers: Indicates the sum of the read/write per second of all disks attached to the ECS instance. The group data graph displays the average value of the data of all ECS instances in the application group.

  4. Set the time range.

    You can set the time range to “Half an Hour”, “Six Hours”, “One Day”, or “One Week”.

    • Half an Hour: Collects monitored data in last half an hour by default when you log on to the infrastructure monitoring page. In this statistical period, data is collected every minute, which is the finest query granularity by EDAS.

    • Six Hours: Collects monitored data in last six hours. In this statistical period, data is collected every five minutes.

    • One Day: Collects monitored data in last 24 hours. In this statistical period, data is collected every 15 minutes.

    • One Week: Collects monitored data in last seven days. In this statistical period, data is collected every one hour, which is the longest statistical cycle provided by EDAS.

    Note:

    Start Time and End Time on the page indicate the time span of the current view. When you set one of them, the other one is automatically updated. For example, if you select “30 minutes” and set “End Time” to “2016-05-20 12:00:00”, “Start Time” automatically changes to “2016-05-20 11:30:00”.

    Monitor data is automatically updated based on the selected interval.

  5. (Optional) Click “Zoom In” under a metric to view the enlarged graph of the metric, and adjust the time range in the enlarged graph.

Thank you! We've received your feedback.