Analysis: Azure Recommendations

Analysis: Azure Recommendations

#340510

Description

The /analysis/cloud/azure/<analysisId>/results resource is used to return a collection of Azure Virtual Machine instance recommendations after optimization analysis has been performed on your collected Azure infrastructure data.

For each instance recommendation, you can also download a PDF version of the Impact Analysis and Recommendation Report, which details system impact based on the recommendations. This report is useful to application owners who need to review recommended changes to their application resources. See the rptHref resource element for details on how to download this report.

To return a collection of all instances included in an Azure optimization analysis, see Analysis: Azure Systems.

Resource

/analysis/cloud/azure/<analysisId>/results

Supported Operations

Table: Azure Recommendations Supported Operations

HTTP Method

Input

Output

Description

GET /analysis/cloud/azure/ <analysisId>/results

Path Parameter:

Query String Parameter Options:

Accept:

  • application/json

Collection of (JSON):

Returns a collection of recommendations for an Azure analysis.

Specify application/json in the request header for returned recommendations in JSON format.

Note: The returned recommendations can be in either JSON or Terraform-map format.

Example: Returning Azure Instances with Terminate Recommendations

GET /analysis/cloud/azure/ <analysisId>/results

Path Parameter:

Query String Parameter Options:

Accept:

  • application/terraform-map

Collection of (Terraform-map):

Returns a collection of recommendations for an Azure analysis.

Specify application/terraform-map in the request header for returned recommendations in Terraform-map format.

Note: The returned recommendations can be in either JSON or Terraform-map format.

Example: Returning Azure Recommendations with Low Effort in Terraform-map Form

Parameters

Path Parameters

Table: Azure Recommendations Path Parameters

Parameter Name

Type

Description

analysisId

string

The unique referenced ID of the Azure analysis.

Query String Parameters

Table: Azure Recommendation Query String Parameters

Parameter Name

Type

Description

Element Filters

string

You can use element filters to return a targeted subset of the recommendations. See the "F" (filter) designation in the Response schema table for a list of elements that support filtering. Refer to Filters for a complete description of this common operation feature.

Usage example:

../results?recommendationType=Upsize

includeAttributes

  • true
  • false

Indicate whether or not to return system attributes:

  • true—returns all the system attributes;
  • false—(default) suppress system attributes from the response output.

Usage example:

../results?includeAttributes=true

dataQuality

string

Allows you to indicate which workload type data collection detail to return for the system. See dataQuality for the returned details.

Specify a quoted list of workload type names for which data to be returned, separated by commas. The supported workload type names can be found in the Data Center Explorer (DCE) from the Analysis Console. See DCE Virtual Environment Workload Viewer to find a list of the supported workload types from the Analysis Console.

Usage example:

../results?dataQuality="CPU Utilization (CINT2006 Rate),Memory Utilization in Percent"

Note: A "400-Bad Request" error message is returned if a non-supported workload type is specified.

Response

Table: Azure Recommendations Response Schema

Element

Type

Filter/Sort

Description

entityId

string

F

The Densify assigned entity ID of the cloud system.

resourceId

string

F

The Azure identifier assigned to the instance system.

accountIdRef

string

F

The Azure subscription identifier.

currentType

string

F

The current instance type of the Azure system.

recommendationType

string

F

The recommended action for the system.

This is also known as the Optimization Type in the Densify Console (see Optimization Type Color-Coding Summary).

The following types of recommended actions are supported for this cloud platform:

  • "Just Right"—this instance is optimally sized for the workload;
  • "Upsize - Optimal Family"—this instance should be upsized to a more optimal instance family;
  • "Upsize"—this instance should be upsized to an instance within the same instance family;
  • "Terminate"—this instance should be terminated;
  • "Downsize - Optimal Family"—this instance should be downsized to an instance belonging to a more suited instance family;
  • "Downsize"—this instance should be downsized to an instance within the same instance family;
  • "Modernize - Optimal Family"—this instance should be modernized to an instance belonging to a more optimal instance family;
  • "Modernize"—this instance should be modernized to an instance within the same instance family.
  • "Not Analyzed"—this instance has no recommendation due to insufficient workload information.

Systems with insufficient information for analysis do not have a recommendation and are returned with limited elements. See Example: Returning Azure Systems with No Recommendations.

recommendedType

string

F

The recommended instance type after Densify optimization analysis.

implementationMethod

string

F

[Self Optimization|Manual|N/A]

Specifies whether this system is configured for Self-Optimization or Manual actioning based on the recommended action (recommendationType) and on the Self-Optimizing Automation policies.

predictedUptime

percentage

 

The predicted uptime (%) for the system is based on the percentage of hours CPU utilization data is present in the historical interval specified in the policy settings. Predicted uptime % for new systems started mid-way within the historical interval is calculated from the time/date that the system was started as opposed to the beginning of the interval resulting, in more accurate prediction for the future.

totalHoursRunning

string

 

The total uptime hours for the system.

name

string

F

The Azure name assigned to the system.

rptHref

string

 

The reference resource to the Impact Analysis and Recommendation Report (also known as the Application Owner report). The PDF report for the specified instance is available for download after the reporting database tables have been updated (i.e. after RDB populate has been executed). By default, the reporting database tables are updated once every night.

See Viewing the Impact Analysis and Recommendation Report for details on the content of the report.

Use Accept : application/octet-stream in the request header to download the PDF file.

Example: Downloading an Impact Analysis and Recommendation Report

approvalType

string

F

The approval setting for the system recommendation.

The value in this string is derived from the Self-Optimizing Automation policies in implementationMethod and the approval attribute attr_ApprovalSetting.

Possible settings include:

  • "na"— not approved;
  • "all"—approve any change;
  • "<recommended-instance-type>"—approve changing the instance to the specified <recommended-instance-type>.

densifyPolicy

string

F

The Densify policy used for optimization analysis.

savingsEstimate

string

 

The difference between the current and recommended instance type cost (i.e. catalog cost), with predicted uptime taken into consideration (i.e. [currentCostrecommendedCost] * predictedUptime ).

effortEstimate

string

F

[Very High|High|Medium|Moderate|Low|Very Low|None|Impossible]

This element describes the effort required to investigate and implement the Densify recommendations. Effort for each system is calculated by rule-driven analytics based on factors (such as instance family change, storage change, data quality checks, feature changes, etc.) that can be configured in the policy settings and rule set which captures best practices.

"Impossible" effort is a result of a manual override for the instance.

powerState

string

F

The power state of the system.

recommendedHostEntityId

string

F

The Densify entity ID of the catalog instance for the recommended instance type.

currentCost

string

 

The cost of the existing instance type (i.e. the instance catalog cost).

recommendedCost

string

 

The cost of the recommended instance type (i.e. the instance catalog cost) after Densify optimization analysis.

serviceType

string

F

The Azure cloud service type:

  • Virtual Machine

currentHourlyRate

string

 

The hourly rate for the current instance type (i.e. instance catalog cost / monthly hours). This value is rounded to the nearest penny.

recommendedHourlyRate

string

 

The hourly rate for the recommended instance type (i.e. instance catalog cost / monthly hours). This value is rounded to the nearest penny.

attributes

array of

  • id
  • name
  • value

 

System attributes are properties set during the data collection process by a vendor platform (i.e. Azure) or by Densify for analytics.

Note: The attribute array is only returned when includeAttributes=true is added to the query string.

provisioningId

string

 

This element is used to identify the terraform-map recommendations for a unique system and corresponds to the "Provisioning Id" Azure user-defined resource tag value. The "Provisioning Id" resource tag is used to uniquely identify a system, since its resourceId or system name could possibly change after an instance type update.

See Example: Returning Azure Recommendations with Low Effort in Terraform-map Form

If the "Provisioning Id" user-defined resource tag value is not set, then the system name is used to identify the recommendations.

recommFirstSeen

Unix time

(in milliseconds)

 

The first date and time the recommended instance type (i.e. the recommendedType element) was provided by Densify (Unix Epoch time, in milliseconds).

recommLastSeen

Unix time

(in milliseconds)

 

The latest date and time the recommended instance type (i.e. the recommendedType element) was provided by Densify (Unix Epoch time, in milliseconds).

recommSeenCount

integer

 

The number of times Densify suggested the recommended instance type (i.e. the recommendedType element). This is the count of Densify analysis processes which produced the same recommended instance type from recommFirstSeen to recommLastSeen. This value is updated when the RDB populate task (i.e. the reporting database update process) is executed. The RDB populate task compares the current recommendedType with the new recommendedType to update the recommSeenCount counter.

Note:  In typical production environments, where the RDB populate task is scheduled to run once daily post data collection and analysis, the recommSeenCount value will reflect exactly the number of times the recommended instance type was provided. If your environment executes the RDB populate task more than once daily, the recommSeenCount value will be inflated beyond the actual number of times the instance type was recommended. Contact support@Densify.com if you have concerns about the recommSeenCount value.

auditInfo

  • dataCollection:
    • dateFirstAudited
    • dateLastAudited
    • auditCount
  • workloadDataLast30:
    • firstDate
    • lastDate
    • totalDays
    • seenDays

 

The following system data collection details are returned:

  • dateFirstAudited—the first time data was collected for this system (Unix epoch time, in milliseconds);
  • dateLastAudited—the most recent data collection time (Unix epoch time, in milliseconds);
  • auditCount—the number of times that data was collected.

The following system workload collection details, for the last 30 days, are returned:

  • firstDate—the first time workload data was collected for this system (Unix epoch time, in milliseconds);
  • lastDate—the most recent workload data collection time for this system (Unix epoch time, in milliseconds);
  • totalDays—the difference in days between firstDate and lastDate;
  • seenDays—the number of days that at least one workload data was added into Densify for this system.

Note: If no workload data is collected for the system in the last 30 days, then the workloadDataLast30 element block is not returned.

Note: The values in auditInfo are updated once a day, after the data collection and RDB populate processes are complete (i.e. the reporting tables have been updated with latest data collected).

dataQuality

array of:

  • workloadName
  • firstSeen
  • lastSeen
  • completeDays
  • partialDays

 

The dataQuality array provides workload type data collection details for the system:

  • workloadName—the workload type name (see DCE Virtual Environment Workload Viewer to find a list of the supported workload types from the Analysis Console);
  • firstSeen—the first time this workload was collected (Unix Epoch time, in milliseconds);
  • lastSeen—the most recent time this workload was collected (Unix Epoch time, in milliseconds);
  • completeDays—the number of complete days that this workload data was collected;
  • partialDays—the number of partial days that this workload data was collected.

Note: The dataQuality array is only returned when the dataQuality query string is specified in the request. In addition, if you specify a workload type in the request, for which system workload data does not exist in Densify, then no data is returned.

Note: The values in the dataQuality array are updated after data collection and subsequent updates to the reporting tables (i.e. RDB Populate process) are completed.

Examples

Example: Returning Azure Instances with Terminate Recommendations

The following example shows you how to return a collection of Azure instances with "Terminate" recommendations.

Example: Returning Azure Recommendations with Low Effort in Terraform-map Form

The following example shows you how to return a collection of recommendations with Low effort in terraform-map form. The label of each recommendation (i.e. "st01-dev-west-108", "st01-dev-west-110" in the example below) is the provisioningId element:  "Provisioning Id" Azure user-defined resource tag or system name value.

Example: Downloading an Impact Analysis and Recommendation Report

The following example shows you how to download a PDF Impact Analysis and Recommendation Report from the rptHref resource element provided in the instance recommendation output.

Note: HTTPS needs to be enabled to download the Impact Analysis and Recommendation Report PDF.

Example: Returning Azure Systems with No Recommendations

The following example shows you how to return a collection of Azure systems without recommendations. These systems typically do not have adequate data for optimization analysis and have the "Not Analyzed" designation in the recommendationType element.