Ideal age vector based file retention in a software testing system

Abstract

Embodiments relate to ideal age vector based file retention in a software testing system. An aspect includes determining an ideal age vector of historical bug age data corresponding to a software product in the software testing system, wherein the software testing system stores a plurality of version files of the software product. Another aspect includes based on receiving a new version file of the software product, determining a file to delete of the plurality of version files and the new version file based on the ideal age vector. Yet another aspect includes deleting the determined file to delete such that a number of remaining version files is equal to a number of the plurality of version files.

What is claimed is: 
     
         1 . A computer implemented method for ideal age vector based file retention in a software testing system, the method comprising:
 determining an ideal age vector of historical bug age data corresponding to a software product in the software testing system, wherein the software testing system stores a plurality of version files of the software product;   based on receiving a new version file of the software product, determining a file to delete of the plurality of version files and the new version file based on the ideal age vector; and   deleting the determined file to delete such that a number of remaining version files is equal to a number of the plurality of version files.   
     
     
         2 . The method of  claim 1 , further comprising determining the ideal age vector based on a probability distribution function of historical bug age data. 
     
     
         3 . The method of  claim 1 , wherein determining the file to delete of the plurality of version files and the new version file based on the probability distribution function comprises:
 for each of a plurality of deletion candidate version files, determining an actual age vector for remaining version files without the deletion candidate;   determining an actual age vector having a smallest distance from the ideal age vector; and   selecting the deletion candidate version file corresponding to the actual age vector having the smallest distance from the ideal age vector as the file to delete;   wherein a respective actual age vector is determined using each of the new version file and the plurality of version files as the deletion candidate.   
     
     
         4 . The method of  claim 3 , wherein the distance is determined based on an absolute value of a difference between the actual age vector components and the ideal age vector components. 
     
     
         5 . The method of  claim 3 , wherein the distance is determined based on a component-wise nearest neighbor distance. 
     
     
         6 . The method of  claim 3 , wherein the distance is determined based on a Euclidean distance. 
     
     
         7 . The method of  claim 3 , further comprising determining an age of an identified bug in the software testing system using the remaining version files; and
 updating the probability distribution function using the determined age.   
     
     
         8 . The method of  claim 1 , further comprising:
 determining that an amount of storage required by the remaining version files is greater than an amount of available storage; and   based on determining that the amount of storage required by the remaining version files is greater than the amount of available storage, determining an additional file to delete of the remaining version files based on the probability distribution function; and   deleting the determined additional file.