Intelligent application diagnostics

A method for application diagnostics of processes running on a user device (UD), the method comprising:

• providing, by a probe application, timing information to an operating system of the UD, wherein the timing information provides instructions for the operating system to wake up the probe application from a sleep mode every first time period and every second time period;
• upon waking from the sleep mode every first time period:
• sending, by the probe application, a request to the UD to determine key performance indicators (KPI's) of a plurality of processes running on the UD;
• determining, by the probe application, a top predetermined number (N) processes of the plurality of processes that are using the most computing resources, based on the KPI's;
• discarding the KPI's of the processes that are not in the top N processes;
• storing filtered metrics based on the KPI's of the top N processes in a local memory;
• and
• entering, by the probe application, into the sleep mode;
• and
• upon waking from the sleep mode every second time period:
• retrieving, by the probe application, the filtered metrics of every iteration of the first time period from the local memory;
• sending a data package comprising the filtered metrics to a remote monitoring server for analysis and storage;
• removing the filtered metrics from the local memory;
• and
• entering, by the probe application, into the sleep mode.

A non-transitory computer-readable medium having stored thereon a plurality of sequences of instructions which, when executed by one or more processors of a user device (UD), cause the one or more processors to direct a probe application to:

• provide timing information to an operating system of the UD, wherein the timing information provides instructions for the operating system to wake up the probe application from a sleep mode every first time period and every second time period;
• upon waking from the sleep mode every first time period:
• send a request to the UD to determine key performance indicators (KPI's) of a plurality of processes running on the UD;
• determine a top predetermined number (N) processes of the plurality of processes that are using the most resources of the UD, based on the KPI's;
• discard the KPI's of processes of the plurality of processes that are not in the top N processes;
• and
• determine a universal consumption factor (UCF) for each process in the top N processes based on the KPI's of each process, wherein the UCF provides an aggregate performance assessment of a respective process running on the UD based on:
• (i) a process CPU time that the respective process uses the one or more processors during the first time period, (ii) a UD CPU time that the UD uses the one or more processors during the first time period, (iii) a process bytes of a total data sent and received by the respective process during the first time period, and (iv) a UD bytes of a total data sent and received by the UD during the first time period;
• and
• store filtered metrics based on the KPI's and the UCFs of the top N processes in a storage device of the UD;
• and
• upon waking from the sleep mode every second time period:
• retrieve the stored filtered metrics from the storage device;
• send a data package comprising the filtered metrics to a remote monitoring server for analysis and storage;
• and
• remove the filtered metrics from the storage device.