Page tree
Skip to end of metadata
Go to start of metadata
Summary: The sections below explain how the Technical Debt value is calculated and how the default values can be modified if necessary. Technical Debt configuration is undertaken in the CAST Management Studio, via the Assessment Models editor. Note that an extension exists to calculate OMG Technical Debt in addition to Technical Debt described below.

Calculation

Technical Debt is calculated on an Application basis, as follows:

Note that low, medium and high severity violations refer to the weight of a violated Quality Rule as defined in the Assessment Model:

  • Low severity = Weight of 1, 2 or 3
  • Medium severity = Weight of 4, 5 or 6
  • High severity = Weight of 7, 8 or 9

Total Technical Debt per Application

( (% of low severity violations to be fixed  X  # of low severity violations in Application) X
(Weighted time, in hours, for fixing low severity violations) +
(% of medium severity violations to be fixed  X # of medium severity violations in Application) X
(Weighted time, in hours, for fixing medium severity violations) +
(% of high severity violations to be fixed  X  # of high severity violations in Application) X
(Weighted time, in hours, for fixing high severity violations) ) X
Cost per staff hour to fix violations

Technical Debt Added in Current Release per Application

( (% of low severity violations to be fixed  X  # of low severity violations added in current release of Application) X
(Weighted time, in hours, for fixing low severity violations) +
(% of medium severity violations to be fixed  X  # of medium severity violations added in current release of Application) X
(Weighted time, in hours, for fixing medium severity violations) +
(% of high severity violations to be fixed  X # of high severity violations added in current release of Application) X
(Weighted time, in hours, for fixing high severity violations) ) X
Cost per staff hour to fix violations 

About Added violations

Added Violations with low/medium/high severity are violations that:

  1. involve added and updated objects
  2. are not present in the Application in the previous Snapshot
  3. involve a Quality Rule with the appropriate Weight

This means that:

  • some configuration changes can cause the number of Violations of a given Severity to change without "Adding Violations". E.g.: Changing the Weight of a Quality Rule between Snapshots can cause a change in the number of low/medium/high severity violations without any specific "Added Violation"
  • some Objects with Violations but without a relevant checksum (i.e. a value of 0) do not show in the Technical Debt Added (note that checksum values are calculated by CAST AIP for objects resulting from an analysis and are used to determine whether an object has changed between successive snapshots, however, CAST AIP is not able to determine the checksum for certain objects and these objects always have a checksum value of 0).

In these cases, Technical Debt amount in new Snapshot is different from Technical Debt amount in previous Snapshot + Technical Debt Added amount - Technical Debt Removed amount

Technical Debt Removed in Current Release per Application

( (% of low severity violations to be fixed  X  # of low severity violations removed in current release of Application) X
(Weighted time, in hours, for fixing low severity violations) +
(% of medium severity violations to be fixed  X  # of medium severity violations removed in current release of Application) X
(Weighted time, in hours, for fixing medium severity violations) +
(% of high severity violations to be fixed  X # of high severity violations removed in current release of Application) X
(Weighted time, in hours, for fixing high severity violations) ) X
Cost per staff hour to fix violations

About Removed violations

Removed Violations with low/medium/high severity are violations that:

  1. involve removed and updated Objects
  2. are not present anymore in the Application in the current Snapshot
  3. involve a Quality Rule with the appropriate Weight

This means that

  • some configuration change can cause the number of Violations of a given Severity to change without "Removing Violations". E.g.: Changing the Weight of a Quality Rule between Snapshots can cause a change in the number of low/medium/high severity violations without any "Removed Violation"
  • some Objects with Violations but without a relevant checksum (i.e. a value of 0) do not show in the Technical Debt Removed (note that checksum values are calculated by CAST AIP for objects resulting from an analysis and are used to determine whether an object has changed between successive snapshots, however, CAST AIP is not able to determine the checksum for certain objects and these objects always have a checksum value of 0).
  • Excluded Violations do not show in the Technical Debt Removed

In these cases, Technical Debt amount in new Snapshot is different from Technical Debt amount in previous Snapshot + Technical Debt Added amount - Technical Debt Removed amount

Calculation - variable description

The variables listed in the calculation above are described below.

Please see the section below entitled Modifying Technical Debt calculation variables about modifying any of the variables marked as Configurable = Yes.

Variable Name

Description

Configurable

Default Value

% of low severity violations to be fixed

Only a portion of the low severity violations will be fixed

Yes

0%

# of low severity violations

Actual # of low severity (level 1,2,3) violations across all health factors

No (comes directly from analysis)

Not Applicable

# of low severity violations added in current release

Actual # of low severity (level 1,2,3) violations across all health factors added in current release

No (comes directly from analysis)

Not Applicable

# of low severity violations removed in current release

Actual # of low severity (level 1,2,3) violations across all health factors removed in current release

No (comes directly from analysis)

Not Applicable

% of medium severity violations

Only a portion of the medium severity violations will be fixed

Yes

50%

# of medium severity violations

Actual # of medium severity (level 4,5,6) violations across all health factors

No (comes directly from analysis)

Not Applicable

# of medium severity violations added in current release

Actual # of medium severity (level 4,5,6) violations across all health factors added in current release

No (comes directly from analysis)

Not Applicable

# of medium severity violations removed in current release

Actual # of medium severity (level 4,5,6) violations across all health factors added in current release

No (comes directly from analysis)

Not Applicable

% of high severity violations

Only a portion of the high severity violations will be fixed

Yes

100%

# of high severity violations

Actual # of high severity (level 7,8,9) violations across all health factors

No (comes directly from analysis)

Not Applicable

# of high severity violations added in current release

Actual # of high severity (level 7,8,9) violations across all health factors added in current release

No (comes directly from analysis)

Not Applicable

# of high severity violations removed in current release

Actual # of high severity (level 7,8,9) violations across all health factors added in current release

No (comes directly from analysis)

Not Applicable

Weighted time, in hours, for fixing LOW severity violation

Not all violations will need the same amount of time, hence we take the weighted time to fix the violations.

Weighted based on the distribution of level of difficulty to fix violations. Violations will be categorized as follows:

- Easy
- Hard
- Very Hard

Wt. time to fix low severity violations can be given a new value using the variables described below:

- (Low_%Easy X Low_Time_Easy) + (Low_%Hard+ X Low_Time_Hard) + (Low_%Very_Hard X Low_Time_Very_Hard)

For example using the variables below:

(0.90 X 0.5) + (0.09 x 1) + (0.01 x 8) = 0.62

Yes


0.62 hours


Low_%Easy = % of violations which are "Easy"

N/A (intermediate computing)

90%


Low_Time_Easy = Time take for fixing "Easy" violations

N/A (intermediate computing)

0.5 hour


Low_%Hard = % of violations which are "Hard"

N/A (intermediate computing)

9%


Low_Time_Hard = Time take for fixing "Hard" violations

N/A (intermediate computing)

1 hour


Low_%Very_Hard = % of violations which are "Very_Hard"

N/A (intermediate computing)

1%


Low_Time_Very_Hard = Time take for fixing "Very_Hard" violations

N/A (intermediate computing)

8 hours

Weighted time, in hours, for fixing MEDIUM severity violation

Not all violations will need the same amount of time, hence we take the weighted time to fix the violations.

Weighted based on the distribution of level of difficulty to fix violations. Violations will be categorized as follows:

- Easy
- Hard
- Very Hard

Wt. time to fix medium severity violations can be given a new value using the variables described below:

- (Medium_%Easy X Medium_Time_Easy) + (Medium_%Hard X Medium_Time_Hard) + (Medium_%Very_Hard X Medium_Time_Very_Hard)

For example using the variables below:

(0.90 X 0.5) + (0.09 x 4) + (0.01 x 16) = 0.97

Yes

0.97 hours


Medium_%Easy = % of violations which are "Easy"

N/A (intermediate computing)

90%


Medium_Time_Easy = Time take for fixing "Easy" violations

N/A (intermediate computing)

0.5 hour


Medium _%Hard = % of violations which are "Hard"

N/A (intermediate computing)

9%


Medium _Time_Hard = Time take for fixing "Hard" violations

N/A (intermediate computing)

4 hour


Medium _%Very_Hard = % of violations which are "Very_Hard"

N/A (intermediate computing)

1%


Medium _Time_Very_Hard = Time take for fixing "Very_Hard" violations

N/A (intermediate computing)

16 hours

Weighted time, in hours, for fixing HIGH severity violation

Not all violations will need the same amount of time, hence we take the weighted time to fix the violations.

Weighted based on the distribution of level of difficulty to fix violations. Violations will be categorized as follows:

- Easy
- Hard
- Very Hard

Wt. time to fix high severity violations can be given a new value using the variables described below:

- (High_%Easy X High _Time_Easy) + (High _%Hard X High _Time_Hard) + (High _%Very_Hard X High _Time_Very_Hard)

(0.80 X 1) + (0.19 x 8) + (0.01 x 24) = 2.56

Yes

2.56 hours


High _%Easy = % of violations which are "Easy"

N/A (intermediate computing)

80%


High _Time_Easy = Time take for fixing "Easy" violations

N/A (intermediate computing)

1 hour


High _%Hard = % of violations which are "Hard"

N/A (intermediate computing)

19%


High _Time_Hard = Time take for fixing "Hard" violations

N/A (intermediate computing)

8 hours


High _%Very_Hard = % of violations which are "Very_Hard"

N/A (intermediate computing)

1%


High _Time_Very_Hard = Time take for fixing "Very_Hard"violations

N/A (intermediate computing)

24 hours

Cost per hour of developer time

Blended rate of different people who may work on a violation (architect, lead, developer,QA resource etc.)

Yes

$75/hr

Modifying Technical Debt calculation variables

To modify any of the variables listed in the section above, please use the CAST Management Studio:

  • Open the Assessment Model provided by default in the Assessment Models editor (note that this Assessment Model is the default Model compatible with the current version of AIP Core - i.e. it does not contain any customization that may have been migrated from previous versions of CAST) and ensure you are working in Expert Mode:

Note that if you do decide to use this default Assessment Model, you must ensure that your CAST Dashboard Service is configured to use it.
  • Click the Sizing Model tab and scroll down to the very bottom of the list of Sizing Measures:

  • Select the Technical Debt measure and drill down into the Parameters tab. This will show you the parameters you can modify:

  • Note the Parameter you would like to modify and then move into the Contextual Parameters tab.
  • Locate the Parameter in the list and select it - this will display the Default Value as highlighted below:

  • Modify the values in the Default Value field.
  • Repeat for each Technical Debt Parameter you want to modify.
  • You will then need to generate a new snapshot so that the new values are taken into consideration.

Known limitations

Technical Debt values at functional module level when objects belong to multiple technologies

When a physical source file is detected by CAST as belonging to multiple technologies, the resulting objects will be counted for EACH technology that they belong to and therefore the Technical Debt value for these objects will be incorrect at module level. Take for example a .yml file which may be analyzed as PHP and also as HTML5: for the purposes of Technical Debt, any resulting objects will be counted once for PHP and once for HTML5, therefore creating incorrect values for Technical Debt.

  • No labels