| CPC G06F 11/0787 (2013.01) [G06F 11/0781 (2013.01)] | 20 Claims |

|
1. A method for managing log retention period of an application, the method comprising:
inferring dependencies among applications executing on a distributed system, wherein the applications are operatively connected through a network;
inferring connectivity among the applications executing on the distributed system;
generating an application chain (AC) based on the inferred dependencies and the inferred connectivity;
analyzing distributed logs and the AC to extract health parameters comprising an error count per request type, an error occurrence duration of an error, and a priority level of each application;
deriving a failure score of the AC based on the health parameters and the inferred dependencies among the applications;
inferring at least a user access frequency (UAF) of each application log (AL) and remaining assigned storage space (RASS) for each AL based on the distributed logs;
analyzing the failure score of the AC, the UAF of each AL, and the RASS for each AL;
making, based on the analyzing of the failure score of the AC, the UAF of each AL, and the RASS for each AL, a first determination that the failure score of the AC is greater than a predetermined failure score;
making, based on the first determination, a second determination that the AC comprises a high priority application (HPA);
making, based on the second determination, a third determination that the RASS for the HPA's AL is less than a predetermined storage space limit;
notifying, based on the third determination, an administrator to initiate obtaining of extra storage space prior to extending a current AL retention period of the HPA's AL; and
extending, upon obtaining the extra storage space, the current AL retention period of the HPA's AL to a second AL retention period.
|