Detecting and predicting application performance

A system comprising:

• a memory having computer readable instructions;
• and
• one or more processors for executing the computer readable instructions, the computer readable instructions controlling the one or more processors to perform operations comprising:
• receiving a current source code and a revised source code comprising new code;
• splitting the new code into segments comprising three words;
• creating a set of summarized features by labeling, using a first machine learning model, each segment as a feature, a condition, or an ignored segment based on a value of the second word in the respective segment, wherein values associated with moving, performing, executing, determining equalities, and sending are labeled as features, wherein values associated with if, end-if, and else statements are labeled as conditions, and wherein all other values are labeled as ignored segments;
• assigning a performance indicator type to each summarized feature of the first machine learning model, the performance indicator type comprising one of a response time and a CPU time;
• determining, using a second machine learning model for the assigned performance indicator type and historical data, a value of the performance indicator assigned to each respective summarized feature;
• and
• providing an estimated performance impact of the new code of the revised source code based on the value of the performance indicator.

A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform operations comprising:

• receiving a current source code and a revised source code comprising new code;
• splitting the new code into segments comprising three words;
• creating a set of summarized features by labeling, using a first machine learning model, each segment as a feature, a condition, or an ignored segment based on a value of the second word in the respective segment, wherein values associated with moving, performing, executing, determining equalities, and sending are labeled as features, wherein values associated with if, end-if, and else statements are labeled as conditions, and wherein all other values are labeled as ignored segments;
• assigning a performance indicator type to each summarized feature of the first machine learning model, the performance indicator type comprising one of a response time and a CPU time;
• determining, using a second machine learning model for the assigned performance indicator type and historical data, a value of the performance indicator assigned to each respective summarized feature;
• and
• providing an estimated performance impact of the new code of the revised source code based on the value of the performance indicator.