Available Reports

Overview

This document lists all the standard reports currently available in Capacity Planner along with their description and examples. Each report includes parameters to add flexibility. These allow you to select and filter data of interest and to choose output formats. For more details see, Report Parameters.

If you require a custom report to be added to your project, please contact your Capacity Planner representative.

AWS cost analysis

This report allows you look into what is happening in your estate in different periods of time.

The report includes the following information:

  • You can break down information by tags.
  • You can look at when machines have been idle, and how much it was costing you .
  • Forecast based on current spend.
  • The amount of money spent for a particular product on a particular day.

For more information on configuring cloud cost optimisation recommendations, see Recommendations for cloud environments.

AWS exec report

This report presents an executive summary of all your AWS related costs. It has a minimum amount of parameters, but presents a very high level picture of the costs.

The report includes the following information:

  • Daily trend of costs across all services and accounts.
  • Forecast of where your costs will be by the end of the month.
  • A breakdown of potential savings.
  • Comparisons of service growth.
  • How various services contribute to the final cost.
  • Idle costs
  • Detailed recommendations.

For more information on configuring cloud cost optimisation recommendations, see Recommendations for cloud environments.

Highly inactive VMs

This report identifies virtual machines across the selected baseline model that are considered highly inactive (have very low levels of CPU utilization). Peak, 99th percentile, 95th percentile and mean values are given for each VM and by default the reports are ordered by peak CPU utilisation (from smallest to largest). A summary of the disk space used by each VM is also provided.

This report allows you to very quickly identify the VMs that can be switched off or ones that spend significant amount of time inactive, with very small times of peak activity.

Example

In the image below, you can see an example of an output of this report.

The default sorting can be changed by clicking the arrows above each column to sort by that column.

Cluster events

This report displays all current projected cluster capacity events and the time series for these events. It also provides information on how often this cluster event has been raised and at which severity. You can view the history of each event using the Drill through to Event History option which takes you to the Cluster event history report.

Example

In the image below, you can see an example of an output of this report.

This example shows increasing CPU over time. If this trend continues, the cluster will soon have less than one redundant server available.

Cluster event history

This report shows the event history for each cluster:

  • Blue indicator marks the expected saturation date when the event was first raised.
  • Red indicator marks the most recently calculated saturation date.
  • A series of indicators show each date between the first and the most recent saturation date.

This can be used to gain an understanding of whether an event is coming closer and more imminent, or moving further into the future as a result of a flattening trend.

A chart is also provided showing the time series of allocated vCPUs in the cluster, giving an indication of growth that may also contribute to changes in the cluster saturation dates.

Example

In the image below, you can see an example of an output of this report.

This example shows a cluster event that has been raised a number of times. When first raised, the projected saturation date was 29 October 2018. The most recent saturation date predicted is now 18 September as indicated by the red indicator on the chart.

The lower chart shows a time series trend of allocated vCPUs in this cluster. There has been a rise over the course of the last 3 months which is most likely leading to the change in projected saturation date for the cluster.

Data coverage

This report allows you to monitor the volume of data received each day by Capacity Planner across different entity types and metrics.

You can select one or more entity types, and one or more metrics at minimum. The results can be shown at different summary levels, with the ability to group the results by the following:

  • Entity type
  • Entity name
  • Metric name
  • Grouping values

Coverage is shown in a daily matrix report, with the following options available:

  • Calculation — the report can show either number of data points received for each day, or the number of entities supplying data for a given date range that you selected.
  • Display option:
    • The report can display either all entities, or entities with incomplete coverage.
      Incomplete coverage means that Capacity Planner received fewer data points than expected, or that there are gaps in data due to an issue in the system.
    • You can select between full colour and greyscale.
  • Calculation option — as a proportion of the row maximum or as a proportion of the report maximum.
  • Scale type — can be linear or logarithmic.
  • Display Data — you can display or hide detailed data from the report. If you are only interested in observing the colour variations rather than the absolute numbers, select Data Hidden.

Example

In the image below, you can see an example of an output of this report.

Metric analysis — multi-chart

This report allows you to create box plots, bar charts, line charts, and radar charts as well as linear trend charts and scatter plots with linear regression calculations. The report can be run using single or multiple metrics on groups of entities at any aggregation level available within Capacity Planner.

If you want to display all entities on the same chart, use the Metric analysis — single-chart report.

An additional feature allows you to enter your own RAG (Red, Amber, Green) threshold values and display and sort by them. If no threshold operators are selected, the RAG status column is hidden and RAG threshold lines and status column do not appear on the report.

The Linear Trend and Scatter Plots calculate linear regression lines and associated intercept, gradient, and R2 values. Reports can be filtered and ordered by R2 values to allow the most significant ones to be identified. In these reports red and amber threshold values are used to calculate values on the best fit line allowing projected dates and values to be displayed. For example, in the Linear Trend report, the report display shows the projected dates at which amber and red threshold values would be reached.

Available chart types are:

  • Box plot — this chart can be used for daily summaries and above (day, week, month etc.). It shows percentile values as a box with whiskers above and below to indicate the range of value distribution. This is the only chart type that can show more than one percentile value. However, the Percentile/Value selector still controls the metric used for RAG thresholds.
  • Bar chart — values are shown as vertical bars.
  • Line chart — values are plotted as points on a line.
  • Linear trend — this chart plots metric values and linear regression line for any given metric. This chart should be used with hourly average or unaggregated data for shorter term trends. Longer term trends can make use of daily summary values.
  • Radar chart — this chart plots all entities on a single radar visualization. This chart works well for comparing metrics but we recommend to use it with a relatively small number of entities to improve readability.
  • Scatter plot (metric pairs) and scatter plot (metric pairs reversed) — these chart types appear in the drop-down list if two metrics are selected. They calculate linear regression lines based on metric pair values. Switching between these two report types reverses the X and Y axes of the report. Red and Amber threshold values calculate X axis values from the given Y axis values based on the regression line.
  • Stacked bar chart — this chart presents a single bar per date showing the relative metric values.

Example

In the image below, you can see an example of an output of this report.

It shows CPU and memory daily box plots with the red RAG threshold set to 80% and amber threshold set to 70%. The report is ordered by RAG status.

Metric analysis — single-chart

This report allows you to create bar charts, box plots, line charts, and treemaps with one or more entities shown on the same chart. Plotting multiple entities on the same chart can enable easier side by side comparisons.

If you want to display each entity on a separate chart, use the Metric analysis — multi-chart report.

Available chart types are:

  • Bar chart — values are represented as horizontal bars. The entity names, dates, and values can be sorted on the chart using the sort indicator .
  • Box plot — this chart can be used for daily summaries and above (day, week, month etc.). It shows percentile values as a box with whiskers above and below to indicate the range of value distribution. This is the only chart type that can show more than one percentile value. However, the Percentile/Value selector still controls the metric used for RAG thresholds.
  • Line chart — values are plotted as points on a line.
  • Treemap — entities are shown as areas in a rectangle, with size of area representing the value.

Example

In the images below, you can see examples of line charts, box plots and bar charts and for this report.

Line chart report:

Box plots report:

Bar charts report:

Metric threshold analysis

This report allows you to select a statistical measure for any metric and use it in combination with a threshold. The measure is an indication of how much total time in the chosen baseline period that metric has been above the specified threshold. You can select up to 3 metrics for threshold breach checks.

The report is presented in a visual tabular format of horizontal bar charts. A bar is shown as red if a threshold was crossed.

You can drill through the measures and investigate the time series data for the entity/metric combination. To do this, click the bar for the selected entity.

Example

In the image below, you can see an example of an output of this report.

It shows peak CPU, memory and CPU ready% per VCPU over an entire baseline period.

Metric threshold percentiles

This report shows all statistical values for selected entities for a given metric. Statistical values available are: minimum, 5th percentile, 25th percentile, 50th percentile (median), mean, 75th percentile, 95th percentile, 99th percentile, maximum and latest value.

A threshold filter is available to view only those entities that have breached a user-defined threshold. To use this feature, select values for Filter Operator and Filter Threshold. This is particularly useful for identifying very busy or inactive servers. A box plot view is also available by selecting the Chart display type.

You can drill down to further plots and time series for any entity in the table or chart. To do this, click the entity name on the table or the box plot in the chart.

Example

In the image below, you can see an example of an output of this report.

It shows all VMs where the 95th percentile of CPU is greater than a specified value. In this case, using 0 shows all measures for this metric for all machines.

Scatter plot

This report allows you to generate a scatter plot showing metric pairs values from any two entities.

Example

In the image below, you can see an example of an output of this report.

Time above threshold

This report allows you to select a baseline, group of entities, and a metric, and determine the total time spent above a given threshold for each entity. You can also identify the single longest time period for which it was above that threshold. You can filter results further by days of the week and time of day.

You can drill through the length of longest episode for a detailed time series of when that episode occurred.

Example

In the image below, you can see an example of an output of this report.

VM storage summary

This report displays the comparison between storage allocated to a virtual machine and storage used by the virtual machine OS drives. It can be used to identify potential storage waste in an environment. The report is ordered by most wasteful VMs. When the report is generated, you can expand on drive details for more information on the VM drives and utilisation compared to allocated thick storage.

When filtering data that should be displayed in the report, you can specify the VM state, Current or Removed. Current VMs are the ones that are still actively generating data.

For investigative or exploratory work, we recommend using the Current state for reporting.

Example

In the image below, you can see an example of an output of this report.

Service infrastructure

This report allows you to select a grouping of workloads and determine the aggregate long term trend in the CPU and memory utilisation across all workloads, report and server activity and short term operation risks discovered through advanced trending algorithms.

Its purpose is to provide a single report that covers everything needed to present a comprehensive capacity summary for an application, service or any common collection of servers that share a Capacity Planner grouping value.

When you configure the report and select the Service Grouping and Service Grouping Values parameters, the remaining parameters are filled with default values that can be changed.

The report displays a service summary with the following information:

  • A list of idle, overprovisioned and underprovisioned machines along with right-sizing recommendations.
  • When the capacity thresholds will be crossed.
  • Headroom for growth.
  • Short term operational events in order of criticality (high, idle, low).
  • Service risks — resources that require urgent attention.
  • Box plot charts for all servers and trend time series.

When you export the report, you can click through from a server summary on the executive summary page of the report to the detailed box plot representation.

There are a number of components to this report:

  • The executive summary
  • A high-level overview of trends
  • Summary of server activity
  • Summary of server risks
  • Detail on trends
  • Detail on server activity
  • Detail on service risks

In the Example section below, you can see the output of each of these sections.

Report parameters

These are the parameters that can be used with this report:

Parameter Description
Baseline The baseline to be used to generate this report.
Service grouping The name of the grouping used to filter the reporting server scope.
Service grouping value The value of the grouping.
Service subgroup If modelling of a service is made more complex by shared services, this parameter allows for other grouping values to be included in the report.
Service display name The name used at the top of the report to describe the service. If left blank, the Service grouping value is used.
Service tier If a grouping is used to define tiers of an application, it can be defined here. This ensures that each tier is separated in the report and trended separately.
Growth trend model

The model used for calculating long term trends.

Default value: Daily Summary

Growth trend data range type

Defines the point in the data from which the report begins to look back.

Default value: Relative Date Range from latest data

Growth trend relative range

Specifies how far back the report looks.

Default value: Last 'N' Complete Months

Growth trend number of time periods

Specifies how many time periods to use in trending.

Default value: 6. This means that the default trend range is 6 full complete calendar months.

Growth trend red CPU threshold

The level of the long-term trend that would be flagged as red in the trend summary.

Default value: 90%

Growth trend amber CPU threshold

The level of the long-term trend that would be flagged as amber in the trend summary.

Default value: 70%

Growth trend CPU percentile

The daily percentile measure.

Default value: 95th percentile

Growth trend red memory threshold

The level of the long-term trend that would be flagged as red in the trend summary.

Default value: 90%

Growth trend amber memory threshold

The level of the long-term trend that would be flagged as amber in the trend summary.

Default value: 70%

Growth trend memory percentile

The daily percentile measure.

Default value: 99th percentile

Number of days for current usage

Current usage is the average value of the aggregated percentile selected. When you determine the current usage in the summary, this defines the number of days to use.

Default value: 7

Growth trend display option

Determines whether to use aggregate raw values or percentages in trending.

Default value: Percentages

Idle server CPU percentage

The level at which the server is considered idle if CPU percentile falls below it.

Default value: 5

Low server CPU percentage

The level at which the server is considered low activity if CPU percentile falls below it.

Default: 30

High server CPU percentage

The level at which the server is considered high activity if CPU percentile falls below it.

Default: 70

Right size CPU percentile

The percentile value used in right-sizing calculations and recommendations.

Default value: 95th percentile

Right size minimum CPUs

The minimum CPU configuration allowable when recommending new configuration for VMs.

Default value: 1

Right size CPU buffer

The percentage of capacity that must be reserved when right-sizing to ensure that CPU demand percentile is not above the level of capacity buffer.

Default value: 30%

Right size memory percentile

The percentile value used in right-sizing calculations and recommendations.

Default value: 99th

Right size minimum memory (GB)

The minimum memory capacity allowable when recommending new configuration for VMs.

Default value: 1

Right size memory buffer

The percentage of capacity that must be reserved when right-sizing to ensure that Memory demand percentile is not above the level of capacity buffer.

Default value: 20%

Example

Below, you can see examples of an output of this report.

Executive summary

Summary of trends

Summary of server activity

screenshot

Summary of service risks

screenshot

Detail on trends

screenshot

Detail on server activity

screenshot

Detail on service risks

screenshot

screenshot

Other reports

Except for these advanced reports, you can also create Service, Infrastructure, and Scenario reports directly from a Baseline View or Forward Thinking scenario model.

For more information see, Infrastructure and Service Reports and Scenario Reports.