Optimizing Your Containers - ECS Data Tab

Optimizing Your Containers - ECS Data Tab

#380790

The ECS Service Optimization report provides a summary of optimization opportunities and ways to mitigate risk for your ECS environments.

Data is displayed as charts on the Summary tab or as detailed tabular data on the Data tab, covered here. There are two additional tabs in the lower pane: Account Breakdown and Allocation.

Grouping and Filtering Your ECS Data

You can group and filter your data using the following options:

Filter settings defined using any of the above options are persistent between user login sessions, for this page.

When you export the content of a report with filters applied, the filtered set of systems is exported for the Cluster Breakdown, Namespace Breakdown and all of the ECS tabs.

Note: The functionality is not available on the Kubernetes Details tab.

Summary Banner

Key metrics are displayed in a banner at the top of the report.  If you have applied filters the totals displayed in this summary reflect the filtered results. The values, in the Summary bar are not affected by the Group By feature. See Reviewing the Summary Tab for details of the key metrics.

Data Controls

Watch the video for an overview of the data management features:

These tables provide a few options for managing and viewing your data more effectively. The following controls are applicable to the tables on the Data tab and its sub-tabs Account Breakdown and Allocation.

Icon

Description


Allows you to search for any type of data in the displayed columns. i.e. for a specific account, subscription or project. Typically, you will search for items in the first column, but you can search for a value in the other columns.

Click the icon and enter your search string. All matching rows are displayed in the table. You must enter a string of alpha-numeric characters. Wildcards are not supported.

If the data is on multiple pages (i.e. 1 of 5), all pages are searched for the matching string.

Click this icon to export the content of the currently displayed table to .CSV format.


Click this icon to configure the table to display only the columns of interest. The default settings are re-applied when you navigate away from this page.

If there are more than 30 rows in this table, multiple pages will be indicated here. Click the right/left arrows to move to the next/previous page of a multi-page report.

You will see a vertical scroll bar if all 30 rows cannot be displayed.

Sorting Tabular Data

By default the table is sorted by the first column. The first column is customizable using the Group By feature.

  1. Click the column header to sort the table, in ascending order, using the content of the selected column as the sort key.
  2. Click the column header again, to sort the table, in descending order, again using the content of the selected column as the sort key.
  3. Click the column header a third time to clear the sort and return to the default, sorted by Efficiency Rating.
  4. An up/down arrow is displayed in the column header to indicate the column is being used to sort the content of the table.

Sorting is not persistent between pages or login sessions and this feature is only available on the Data tab and its contained tabs. i.e. you can sort the content of the Cluster Breakdown and Allocationstabs.

Tabular Data

The breakdown of the data for your aggregated container environments is displayed on this tab. You can control how the data is displayed in this table using the Group By feature. The Group By setting is used in the first column and data is retrieved and displayed accordingly.

For example, if you wanted to group the data based on Business Unit, Densify displays the analyzed data sorted based on the value of the Business Unit resource tag. See Using the Group By Feature.

The top table provides a breakdown as follows:

Table: ECS Service Optimization - Data Tab Columns

Column(s)

Description

Group By selection: Application |
Business Unit | etc.

The first column is determined by your Group By selection. You can control how the data is displayed in this table using the Group By feature. Your Group By selection is used in the first column and data is retrieved and displayed accordingly. See Using the Group By Feature.

Efficiency Rating

The Efficiency Rating provides an indication of the health of your aggregated environment. The value is indicated in red, when it is greater than 1.

  • A value of 1 indicates that, at an aggregate level, the environment is efficient according to the defined policy settings.
  • A value less than 1 indicates potential savings opportunities due to over-allocated resources.
  • A value greater than 1 indicates potential risks due to under-allocated resources. The value is indicated in red, when it is greater than 1.

By default, the table is sorted by the Efficiency Rating, in descending order. You can sort the content of the table by any column, by clicking on the column header.

Accounts

This is the number of AWS accounts that are associated with the Application | Business Unit | etc. that is listed in the first column.

Regions

These are the regions that are currently being used to run the selected ECS service.

Clusters

The total number of clusters on which the Application | Business Unit | etc. is running. The grouping is defined in the first column. You can change the Grouping options using the Group By feature.

Tasks

The total number of tasks running in the selected ECS service.

Services

This is the number of unique services on which the Application | Business Unit | etc. is running.

For example, if the data is grouped by application, then the total number of containers running the specified application is listed here.

The number of instances listed in this column/row is a hyperlink that takes you to the ECS Optimization Details > Analysis tab, for the selected grouping. i.e. Only the specific systems, as indicated by the count in the originating table cell, are shown on the Analysis tab. See Optimizing Your Containers - Analysis Tab.

Optimal
Identified Risk
Savings Opportunity
Risk and Savings
Size from Unspecified

These 5 columns provide a breakdown of your ECS services by recommendation type.

  • Optimal—These containers are currently sized correctly and there are no cost savings for this group.
  • Identified Risk—This is the number of containers that are at risk. Containers are identified to be at risk if there is an upsize recommendation.
  • Savings opportunity—These containers are oversized. Containers are identified as a savings opportunity if there is a downsize recommendation.
  • Risk and Savings—These containers need to be resized up and down. i.e. upsize memory and downsize the number of CPU cores.
  • Size from Unspecified—These containers need to be resized based on their utilization. No CPU and/or memory requests or limits have been set.

The number of instances listed in this column is a hyperlink that takes you to the Kubernetes Container Optimization > Analysis tab, for the selected grouping. i.e. Only the specific systems, as indicated by the count in the originating table cell, are shown on the Kubernetes Container Optimization > Analysis tab. See Optimizing Your Containers - Analysis Tab.

Values are colour-coded for improved readability. Colour coding is listed in Container Optimization Types.

Percent Optimal (%)

This is the percentage of total containers that are sized correctly. (Optimal Count /Total Systems)

Surplus CPU Request(CPU Units)

This is the number of surplus CPU cores for the selected Application | Business Unit | etc.

Calculated as (Current CPU Request - Recommended CPU Request)

The asterisk on this value indicates that containers with unspecified values are not included in this calculation.

Surplus CPU Request (%)

This is the number of surplus CPU cores for the selected Application | Business Unit | etc. expressed as percentage. Calculated as (Surplus CPU Request/Current CPU Request)*100

The asterisk on this value indicates that containers with unspecified values are not included in this calculation.

Surplus Memory Request (GB)

This is the amount of surplus memory for the selected Application | Business Unit | etc.

38

Calculated as (Current Memory Request - Recommended Memory Request)

The asterisk on this value indicates that containers with unspecified values are not included in this calculation.

Surplus Memory Request (%)

This is the amount of surplus Memory for the selected Application | Business Unit | etc. expressed as percentage. Calculated as (Surplus Memory Request/Current Memory Request)*100

The asterisk on this value indicates that containers with unspecified values are not included in this calculation.

Totals

The totals at the bottom of the table indicate the total for each column, for all pages, if you have a multi-page report.

The number of instances listed in this column/row is a hyperlink that takes you to the ECS Optimization Details > Analysis tab, for the selected grouping. i.e. Only the specific systems, as indicated by the count in the originating table cell, are shown on the Analysis tab. See Optimizing Your Containers - Analysis Tab.

Vertical scroll bars may be displayed to scroll through the entire list, as lower rows may be hidden behind the row of Totals.

Sorting Tabular Data

By default the table is sorted by Efficiency Rating. This is still the case even if you select another Group By option. You can sort the content of this table using any of the available columns, by clicking the column header.

  1. Click the column header to sort the table, in ascending order, using the content of the selected table as the sort key.
  2. Click the column header again, to sort the table, in descending order, again using the content of the selected table as the sort key.
  3. Click the column header a third time to clear the sort and return to the default, sorted by Efficiency Rating.
  4. An up/down arrow is displayed in the column header to indicate the column is being used to sort the content of the table.

Sorting is not persistent between pages or login sessions and this feature is only available on the Data tab and its contained tabs. i.e. you can sort the content of the Account Breakdown and Allocation tabs.

Account Breakdown Tab

The details of the row selected in the top table is expanded and displayed on two tabs on the lower part of the page. The Account Breakdown tab provides a subset of the information from the top of the page, for only the selected row, expanded by account and sorted by account number.

Table: Columns in the Account Breakdown Tab

Column

Description

Account

This is the name of the cluster.

Region

This is the number of unique namespaces in the cluster that are within the grouped selection.

Cluster

This is the number of unique pods within the grouped selection.

Task

This is the number of unique containers based on the grouped selection.

Current CPU Request (CPU Units)

This is the aggregated total current CPU request for the containers in the selected cluster.

Containers with unspecified CPU or memory requests and/or limit values are indicated with a dash (-).

Recommended CPU Request (CPU Units)

This is the aggregated total CPU Request recommended for the containers in the selected cluster.

Surplus CPU Request (CPU Units)

This is the aggregated total CPU request surplus. This value can be negative (at risk) or a positive value.

Current CPU Limit (CPU Units)

This is the aggregated total current specified CPU limit for the containers in the selected cluster.

Containers with unspecified CPU or memory requests and/or limit values are indicated with a dash (-).

Recommended CPU Limit (CPU Units)

This is the aggregated total CPU limit recommended for the containers in the selected cluster.

Surplus CPU Limit (CPU Units)

This is the aggregated total CPU limit surplus for the selected cluster.

This value is calculated as (Current CPU limit - Recommended CPU limit) * Current Count) of the filtered records.

Containers with unspecified CPU or memory requests and/or limit values are indicated with a dash (-).

Current Memory Request (MB)

This is the aggregated total current memory request for the containers in the selected cluster.

Containers with unspecified CPU or memory requests and/or limit values are indicated with a dash (-).

Recommended Memory Request (MB)

This is the aggregated total memory request recommended for the containers in the selected cluster.

Surplus Memory Request (MB)

This is the aggregated total memory request surplus.

This value is calculated as (Current Memory Request - Recommended CPU | Memory Request) * Current Count) of the filtered records.

Containers with unspecified CPU or memory requests and/or limit values are indicated with a dash (-).

Current Memory Limit (MB)

This is the aggregated total current specified memory limit for the containers in the selected cluster.

Containers with unspecified CPU or memory requests and/or limit values are indicated with a dash (-).

Recommended Memory Limit (MB)

This is the aggregated total memory limit recommended for the containers in the selected cluster.

Surplus Memory Limit (MB)

This is the aggregated total surplus memory limit.

This value is calculated as (Current Memory limit - Recommended Memory limit) * Current Count) of the filtered records.

Containers with unspecified CPU or memory requests and/or limit values are indicated with a dash (-).

Services

This is the number of services that reference the specified cluster etc.

The number of instances listed in this column is a hyperlink that takes you to the ECS Service Optimization > Analysis tab, for the selected grouping. i.e. Only the specific systems, as indicated by the count in the originating table cell, are shown on the ECS Service Optimization > Analysis tab. Optimizing Your Containers - Analysis Tab.

Allocation Tab

The allocation details of the row, selected in the top table is expanded and displayed on this tab in the lower part of the page. The Allocation tab provides a subset of the information shown at the top of the page, but only for the selected row.

Based on the row selected in the top table, the following values are calculated:

  • Average—for the selected row the average is calculated as (total / # of services);
  • Maximum—The maximun value of all the items for the selected row;
  • Minimum—The minimun value of all the items for the selected row;
  • Total—The total values of all the items for the selected row.

Calculations only include services where a request or limit value has been specified.

Table: Columns in the Allocation Tab

Column(s)

Description

Current | Recommended CPU Request | Limit

These 4 columns indicate the current and recommended CPU requests and limits.

The average, maximum, minimum and total values are shown for the containers in the selected row.

Surplus CPU Request | Limit

These 2 columns indicate the surplus or deficit CPU requests and limits. If this is an identified risk recommendation the value may be negative indicating that additional resources are required for this cluster.

This value is calculated as (Current vCPU request | limit - Recommended vCPU request | limit) * Current Count) of the filtered records.

Systems with a 'Terminate' recommendation, are excluded from the calculation in the 'Recommended' columns. If all of the systems in the group have a 'Terminate' recommendation, the values in Recommended columns will be indicated as a dash (-).

Containers with unspecified CPU or memory requests and/or limit values are indicated with a dash (-).

Current | Recommended Memory Request | Limit

These 4 columns indicate the current and recommended memory requests and limits.

The average, maximum, minimum and total values are shown for the containers in the selected row.

Surplus Memory Request | Limit

These 2 columns indicate the surplus or deficit memory requests and limits. If this is an identified risk recommendation the value will be negative indicating that additional resources are required for this cluster.

Containers with unspecified CPU or memory requests and/or limit values are indicated with a dash (-).

Predicted Uptime (%)

The predicted uptime (%) for a container is based on the percentage of hours that CPU utilization data is present in the historical interval. The historical interval is specified in the policy settings.

Task Count

The number of tasks.

# of Containers

The number of containers.

Please observe the following for this set of data:

  • Systems with missing data are excluded from the calculations;
  • If there is no data or values are unspecified for a whole group of systems, this is indicated as a dash (-);