Why do we need InSights Inference?
InSights Inference is KPI (Key Performance Indicator) based interface to provide instance of KPIs across various Vectors. It uses Apache Spark 
as job scheduler and executor, and Elasticsearch 
for data.
It is advantageous to get real time observations for each DevOps vector. Also, time series graph could be tracked across each individual vector on the basis of Day/Month/Quarter/Year, resulting in accuracy, and efficiency.
InSights tab with options panel
What's new in InSights with Insights Inference?
- A separate tab to show Observations for each DevOps Vector.
Textual Inferences on the performance (Good or Bad).
The observations are classified into Positive, Negative & Neutral
- Interpretation of various graphs across various vectors.
- Provide information about data at various levels.
Capability to report progress on a Weekly, Monthly and Daily basis.
Configured Spark Jobs are executed on pre-configured intervals for each KPI.
Capability to use existing SDLC Data for applying ML Models.
Capability of a platform to create mathematical formulas for DevOps Metrics.
- Spark Jobs store the result in pre- configured ES Indexes.
The User Interface makes call to a service to display the data in the necessary format.
Minimal Engineering effort is needed to add a new KPI, and report the same – No UI Change is needed.
Current/Default supported KPIs
| KPI Name | DevOps Vector | Mathematical Function |
|---|---|---|
| Average Build Time | BUILD | AVERAGE |
| Total Number of Builds | BUILD | COUNT |
| Number of Successful Builds | BUILD | COUNT |
| Number of Failed Builds | BUILD | COUNT |
| Maximum Build Time | BUILD | MINMAX |
| Minimum Build Time | BUILD | MINMAX |
| Average Complexity | CODEQUALITY | AVERAGE |
| Average Duplicated Blocks | CODEQUALITY | AVERAGE |
| Number of Quality Passed Blocks | CODEQUALITY | COUNT |
| Number of Quality Failed Blocks | CODEQUALITY | COUNT |
| Average Code Coverage | CODEQUALITY | AVERAGE |
| Number of Successful Sonar Executions | CODEQUALITY | COUNT |
| Number of Failed Sonar Executions | CODEQUALITY | COUNT |
| Average Duration of Successful Deployments | DEPLOYMENT | AVERAGE |
| Total Failed Deployments | DEPLOYMENT | COUNT |
| Total Successful Deployments | DEPLOYMENT | COUNT |
| Maximum Deployment Time | DEPLOYMENT | MINMAX |
| Minimum Deployment Time | DEPLOYMENT | MINMAX |
| Number of Defects | DEFECTS | COUNT |
| Number of Closed Defects | DEFECTS | COUNT |
| Number of Open Defects | DEFECTS | COUNT |
| Average Defect Duration | DEFECTS | AVERAGE |
| User with most Commits | DEVELOPMENT | COUNT |