Skip to main content

We've Moved!

Product Documentation has moved to docs.hitachivantara.com
Hitachi Vantara Knowledge

Analyzing performance trends with predictive analytics risk reporting

Use predictive analytics risk reporting tools to predict performance and utilization of infrastructure resources.

Predictive analytics risk reporting overview

With predictive analytics risk reporting, Ops Center Analyzer allows you to generate risk analysis reports based on infrastructure performance trends.

Predictive analytics risk reporting uses several predictive models to calculate the predicted performance trend. Once the predictive analytics license is enabled, the analytics engine uses historical data collected by the Analyzer probe server to calculate performance trends for your risk reports.

Anticipate performance trends

Anticipate performance trends across your infrastructure by generating risk reports from system-defined report definitions, which are available for immediate use.

Use risk reports to report on the affected resources in your system. These resources are at risk of threshold violations. Ops Center Analyzer automatically tracks risk in affected resources and consumers by periodically calculating performance data trends.

NoteThe calculation for selecting affected resources and consumers is executed when the Ops Center Analyzer starts. You can also set the time for these reports to run. The default starting time is 23:23 hours. It might take some time to complete the calculation after a fresh installation.

Generate risk reports on the following:

  • Capacity shortages
  • Capacity planning (growth)
  • Slow response times
  • Anticipated performance bottlenecks
  • Risk for platinum consumers
  • Growing workloads for user and system resources
A screenshot of the Predictive Analytics Report Definitions window.

About predictive risk profiles

Risk profiles specify the target metrics and time span for the predicted performance trend in your risk reports.

Ops Center Analyzer features default risk profiles for immediate use. These system-defined profiles cover a variety of use cases. On Profiles Settings, click on a profile and view the details provided on the Profile Information and Target Metrics tabs.

NoteUsers cannot edit or delete system-defined risk profiles.

The Create Risk Profile function allows you to choose the parameters manually. You can also use the Copy From option to copy an existing profile from the menu, and then modify the settings according to your risk assessment goals. You can edit and delete user-created risk profiles from Profile Settings.

About predictive risk report definitions

Risk report definitions specify the target resources and risk profiles for generating a risk report.

You can select hypervisors, VMs, hosts, FC switches, storage systems, and volumes as target resources for the report. Use Report Definitions to choose a report definition, and view the details on the Report Information, Profile Information, and Target Resources tabs.

NoteUsers cannot edit or delete system-defined report definitions.
You can also create your own risk report definitions. Use the Create Risk Report Definition function to choose the parameters manually. You can also use the Copy From option to copy an existing report definition, and then modify the settings according to your risk assessment goals. You can edit and delete user-created risk report definitions from the Report Definition.

Predictive analytics risk reporting workflow

Calculate performance trend projections by choosing a report definition and adding risk profiles.

In the report definition, select resources for the predictive modeling calculations. Use the risk profile to define the target metrics and time period. The end result is a trend chart with performance trend projections for the selected resources. The reports are similar to the performance-by-metric reports in the Analytics tab. Ops Center Analyzer provides system-defined risk profiles and report definitions for immediate use. Follow the steps in this workflow to begin:
  1. Verify licenses. Check the Ops Center Analyzer Licenses window to verify that the licenses are valid.
  2. Set up risk profiles. These profiles establish the time period and the target metrics for the performance trend projection.
  3. Select a risk definition. This definition instantiates the output for a performance trend projection by combining multiple risk profiles for target resources.
  4. Generate a report. On the Predictive Analytics tab, choose a report definition. Click Predict Result to calculate the trend projection for that report definition.
  5. Analyze the report. On the Report Results tab, choose a report. The selected report opens in a new tab.
  6. Avert the risk. If the performance trend projection shows a potential threshold violation or degraded performance, access the More Actions menu to mark a resource for follow-up analysis, send an email, or run a script. Likewise, you can take corrective action by using available tools in Ops Center Analyzer, such as assigning resources, setting I/O controls, or generating recovery plans.

Adding the Predictive Analytics license

To use risk reporting in Ops Center Analyzer, you must have a valid Predictive Analytics license for each storage system in your environment.

Obtain a Predictive Analytics license from your Hitachi Vantara representative.

Procedure

  1. In the logon window, click Licenses.

  2. Use either of the following methods to register the license:

    • Enter the license key.
    • Specify the license file.
  3. Click Save.

    The license is added to the list.

Creating a predictive risk profile

Create a risk profile for projecting performance trends from 24 hours to 6 months in the future.

Before you begin

You must have a valid Predictive Analytics license to use the predictive analytics risk reporting feature.

Procedure

  1. Click the Administration tab, then select Predictive Analytics Settings Profile Settings.

  2. Click Create Risk Profile.

  3. On the Copy From menu, do one of the following:

    • Select a system-defined or an existing user-defined profile. You can rename it, then modify the parameters according to your risk assessment goals.
    • Select a blank profile (-), the default option in the menu, and construct a new profile by manually selecting the parameters.
  4. Name the profile (required) and provide a description (optional).

  5. Select a time period.

    For example, select 24 hours or 7 days for near-term performance trends. For a long-term projection to determine utilization trends, select 1 month or 6 Months.
  6. In the Target Metrics table, add or remove performance metrics to or from the profile. Click Add Metric, make the following choices, and click OK:

    1. Use the Resource Type menu to display a list of metrics by resource type.

    2. Select target metrics in the Available Metrics field, then click Add. When the Selected Metrics field is complete, click OK.

  7. When the Target Metric table is populated, add a monitoring threshold for each target metric by adjusting the Monitoring setting (ON/OFF) and entering a threshold value.

    Some threshold values have predefined settings, which are adjustable.
  8. Click Save.

Creating risk report definitions

Create risk report definitions to view performance projections for selected target resources.

Before you begin

You must have a valid license to use the Predictive Analytics risk reporting feature.

Procedure

  1. Click the Administration tab, then select Predictive Analytics Report Settings.

  2. On the Copy From menu, do one of the following:

    • Select a system-defined or an existing user-created report definition. Rename it, then modify the parameters according to the risk assessment goals.
    • Select a blank profile (-), the default option, and construct a new profile by manually selecting the parameters.
  3. Name the report definition (required) and provide a description (optional).

  4. Select one or more risk profiles.

  5. Specify the resource scope:

    • To include only selected target resources, choose Selected.
    • To include all resources associated with the selected target resources, choose All Associated.
  6. Click Add Resources.

    1. On the Add Resources window, select Resource Type from the menu.

    2. Select Available Resources and add Selected Resources. You can also remove resources from the list.

    3. Click OK.

  7. Go to the Predictive Analytics tab to generate the report.

Generating a risk report

Specify risk profiles in a report definition to generate performance trend projections.

Before you begin

  • You must have a valid Predictive Analytics license for the predictive analytics risk reporting feature.

Procedure

  1. Generate a report in any of the following ways:

    • From the Predictive Analytics tab, click the Report Definition subtab, select one or more report definitions, then click Predict Result.
    • From the Predictive Analytics tab, click the Search subtab. Select a target resource category, such as consumers, servers, switches, storage systems, or volumes, then enter a resource name. Click Show Predictive Report.
    • From the E2E View window, select one or more consumers and resources, then click Show Predictive Report. The Select Profiles dialog opens.
      1. Specify the Target Resource scope:
        • Selected resources includes only the resources you select.
        • All resources associated with the selected resources includes all associated resources in addition to resources you select.
      2. Select one or more profiles from the Predictive Risk Profile menu, then click OK.
    The generated report opens in a new subtab on the Predictive Analytics tab. View report execution status on the Report Results tab.
  2. On the Report Results subtab, select the generated report to be opened, and click Show Result.

    If you generated a report from the E2E View window or Search subtab, the report name begins with OnDemand followed by the timestamp.
  3. On the left pane of the report, click a metric to view the predicted performance trend.

Understanding performance trend projections in risk reports

You can gain insight into near-term trends and long-term usage of your infrastructure resources from predictive analytics reports.

Report contents

The report is similar to other performance reports in Ops Center Analyzer and includes twin graphs. The left side represents current performance, and the right side represents the predicted performance trend. Zoom in on the graph by clicking and then dragging the cursor over an area. Double-click the graph to return to the normal view.

A screenshot of the predictive analytics report.

Place the cursor on any part of the performance trend to view the timestamp, the performance at that point in time, and the difference between the predicted performance and the threshold value defined for that resource. Because each target metric typically covers multiple resources (selected in the Report Definitions), you can isolate any given resource performance trend by selecting the resource from the list to the right of the graph. To check more resources, click Select Component and add target resources.

Sort by

The report is organized by target metric. View a target metric performance graph by clicking the metric on the left pane. You can select and view multiple graphs.

Sort the list of metrics on the left pane by selecting either Status (default) or Metric. Status organizes by severity of the potential alerts.

Additional actions

When you select resources from the resource list, you can make a note for further tracking. Additionally, click Execute Action to send an email message, run a script, or invoke the Action Template menu.

Report options

Select Save the report as a Report Definition to save the profile and target resource selections.

If conditions have changed or if you have run corrective actions in response to the initial performance trend projection, click Re-run to generate a new report using the same profile and report definitions. Then assess whether your response has averted the risk.

All generated risk reports are listed on the Report Result tab, where they can be sorted, viewed, or deleted.

Enhancing infrastructure management capabilities

You can analyze predicted data in the following contexts to understand performance trends in your infrastructure and allow for better decision making.

Analyzing near-term capacity trends

You can use predictive analytics risk reporting to analyze near-term capacity trends in your infrastructure. Infrastructure administrators know that keeping up with the growing capacity needs of application users is an ongoing effort. By using predictive analytics risk reporting, you can identify capacity needs on a weekly or daily basis by analyzing short-term trend projections. Use the following report definitions to analyze capacity usage trends:

  • Top 5 Pools - Least Available Space
  • Top 5 Pools Consuming Capacity (Free Space)
  • Top 5 System Resources with the Growing Workloads
  • Top 5 User Resources with the Growing Workloads
After you identify the affected storage resources, you can take the following actions:
  • Assign more volumes manually.
  • Create resource assignment rules based on your evaluation of the risk report data.
  • Develop a plan for redistributing storage resources by moving storage pools to different volumes.
Long-term capacity planning

You can plan for future growth by estimating long-term capacity usage trend projections with predictive analytics risk reporting. Capacity planning poses several challenges for infrastructure administrators. Estimates for growth can fluctuate depending on the time of the estimate. When you use predictive analytics risk reporting, you can continually generate estimates for capacity growth. Begin by analyzing near-term trends. Use the following system-defined report definitions to collect data:

  • Top 5 Pools Consuming Capacity
  • Top 5 User Resources with the Busiest Workloads
  • Top 5 System Resources with the Busiest Workloads
  • Top 5 Growing Busy System Resources

Once you get an idea for the trend pattern for the various consumers or application users, create a risk profile that specifies capacity metrics. To maximize the information shown in the trend projection, create one profile for one month, one for three months, and one for six months. Use these profiles in conjunction with report definition that includes consumers.

Near-term performance tuning

You can use predictive analytics risk reporting to fine-tune overall system performance over time to achieve optimization. Because most large-scale IT infrastructures are heterogenous, the task of fine-tuning performance is ongoing. Infrastructure administrators might have the twin goals to make the most of the existing IT equipment (all system resources), while managing resources to achieve stability in performance. With the predictive analytics risk reporting feature in Ops Center Analyzer, you can generate reports with trend projections to anticipate performance fluctuations and adjust resource monitoring.

In this workflow, infrastructure administrators can fine-tune system resource performance and user resource allocation on a weekly or monthly basis.

  1. Set up dynamic thresholds for user and system resources. Use the base dynamic threshold profile or edit it for the following:
    • Metrics: If you want to track performance at a granular level, select more metrics.
    • Plan: You can adjust the profile to monitor during peak times during the week and month
  2. Set up event notifications.
  3. Create risk profiles and report definitions to the corresponding metrics and resources in the profiles.
  4. Generate a risk report along with standard reports to compare past and current trends with the trend projections in the risk report.
  5. Make adjustments in resource allocation, and modify both threshold profiles as needed.
  6. Generate the next round of risk reports and determine if the adjustments averted overutilization of resources.

As an ongoing process, infrastructure administrators have several options on how to act on the information in these risk reports. Ops Center Analyzer offers the following tools to aid in fine-tuning:

  • You can develop a script that is invoked when certain performance events and threshold violations occur, and use the Execute Action function to run that script.
  • You can use Execute Action to invoke a service from Ops Center Automator.
  • If the magnitude of your infrastructure requires thorough analysis of data to avoid service interruptions, you can do the following:
    • To analyze minute-level data, adjust the collection time in Analyzer detail view to track events in finer granularity.
    • To analyze second-level data in Hitachi resources from Ops Center Analyzer, run granular data collection.
Damage Control

You can use the predictive analytics risk reporting feature to evaluate near-term trends and take preventive measures against performance degradation in your infrastructure.

Performance degradation can affect infrastructure with sudden bottlenecks. or worse, extended outages. Consumers might experience I/O problems with lagging response times, an annoying occurrence to application users, and a headache for IT. However, in an extended outage, the application is no longer available for use, causing a work stoppage. This situation can be more than a headache to IT and might result in escalating support calls.

In this workflow description, an infrastructure administrator uses predictive risk reporting in Ops Center Analyzer to respond preemptively to a sudden decrease in performance:

  1. Monitor near-term trends across your infrastructure by using the following report definitions.
    • Top 5 Resources with Worst Response-times and Related Workloads
    • Top 5 Pools - Least Available Space
    • Top 5 User Resources with the Highest Resources
    • Top 5 Platinum Consumers at Risk
  2. When you isolate which resources consistently appear in these reports, create a new risk profile and risk report definition to analyze trend projections for those resources.
  3. Use the new profile and report definition to run risk reports projecting when the performance degradation will occur.
  4. Initiate immediate action to avert performance degradation:
    • Assign resources
    • Set up resource assignment rules
  5. If you determine that the performance trends are recurring, develop a response plan:
    • Adjust thresholds or create a new threshold profile for the resources
    • Set up notifications to track and alert other system administrators
    • Execute a script
    • Run Execute Action
Use risk reporting as an extra layer of monitoring

You can add an extra layer of resource monitoring to give a 10% margin outside your normal threshold limits during day-do-day operations by using predictive analytics risk reporting. To manage performance risks on a daily level, you can generate risk reports to establish a buffer for certain thresholds. Doing so allows infrastructure administrators to predict when thresholds violations will occur. This buffer makes it easier to react to performance problems.

Periodic analysis with predictive analytics risk reporting

You can generate predictive analytics risk reports for the same target resources and metrics to analyze near-term or long-term trends by using risk report definitions periodically.

The following workflow diagram illustrates how to use predictive analytics risk reporting for periodic analysis.

A screenshot of the predictive analytics report.
  1. On Predictive Analytics Report Definition, select a risk report definition, then click Predict Result.
  2. On the Report Results tab, select the risk report and click Show Result.
  3. The predictive analytics risk report opens in a new tab.
Recurring analysis

Place the cursor on any part of the performance trend to view the timestamp, the performance at that point in time, and the difference between the predicted performance and the threshold value defined for that resource. When you select resources from the resource list, you can make a note for further tracking.

A screenshot of the predictive analytics report.

E2E View analysis with predictive analytics risk reporting

You can generate predictive analytics risk reports from the E2E View tab or Search fields to examine the relationship of resources to the target resource based on current conditions.

The following workflow illustrates how to use predictive analytics risk reporting from the E2E View.

Screenshots of launching predictive analytics reports from the end to end view.

  1. On the E2E View tab, select a resource for which you want to view predicted performance data.
  2. When the Select Profiles dialog appears, select a Target Resources option and Predictive Risk Profiles, then click OK.
  3. The predictive analytics risk report opens in a new tab.
Target resource analysis

You can choose to analyze only selected resources from the E2E View or all resources associated with those selected resources.

Preventive actions and predictive analytics risk reporting

You can generate predictive analytics risk reports to check on the effects of corrective or preventive actions by comparing actual performance data with predicted data.

The following workflow diagram illustrates how to use predictive analytics risk reporting for follow-up analysis after taking corrective or preventive actions.

A screenshot that illustrates how to use predictive analytics risk reporting.

  1. On Predictive Analytics Report Definition, select a risk report definition, then click Predict Result.
  2. On the Report Results tab, select the risk report and click Show Result.
  3. The predictive analytics risk report opens in a new tab.
    • Select a resource, then Execute Action to do any of the following:
      • Send an email
      • Run a user-defined script
      • Start a service using Ops Center Automator
    • If conditions have changed or if you have run corrective actions in response to the initial performance trend projection, click Re-run to generate a new report, using the same profile and report definitions.