Method and apparatus for monitoring network servers

Abstract

A server monitoring system includes one or more monitoring engines running on a computer, a system database running on a computer, and a user interface running on a computer to remotely monitor one or more servers across any suitable network and provide alerts and status via either user interface computer or a suitable remote notification system.

1 . A method of running a multi-threaded process from a monitoring server against one or more monitored servers, over a computer network, to obtain status and performance data from the monitored servers, wherein the monitoring server has a CPU and a CPU utilization rate, the method comprising:
 generating one or more monitoring threads, each monitoring thread of the one or more monitoring threads comprising one or more monitoring queries to be executed against one of the one or more monitored servers, the one or more monitoring threads executing asynchronously against the one or more monitored servers, each of the one or more monitoring threads having a preselected duration;   scheduling the one or more monitoring threads to execute against the one or more monitored servers over a preselected time interval;   scheduling each of the one or more monitoring threads to execute its one or more monitoring queries at a preselected execution time within the preselected time interval, such that the preselected execution time of each of the one or more monitoring threads are distributed evenly in relation to each other throughout the duration of the preselected time interval;   returning status and monitoring data that results from the one or more monitoring queries of each of the one or more monitoring threads, to the monitoring server   maintaining a relative timing between each of the one or more monitoring threads in relation to each other throughout one or more iterations of the preselected time interval, wherein the status and performance data returned to the monitoring server and the CPU utilization rate of the monitoring server are continually and evenly regulated in time; and   providing the status and performance data to a user interface means for enabling user monitoring of the one or more monitored servers.   
     
     
         2 . The method of  claim 1  further comprising the steps:
 dynamically creating new monitoring threads, destroying existing monitoring threads as the number of the one or more monitored servers is increased and decreased; and 
 dynamically adjusting the relative timings of all monitoring threads within the preselected time interval to maintain a constant and even regulation in time of the status and performance data to the monitoring server and CPU utilization of the monitoring server when the number of the one or more monitored servers is increased and decreased. 
 
     
     
         3 . The method of  claim 1  further comprising the step:
 dynamically adjusting the relative timings of all of the one or more monitoring threads within the preselected time interval to maintain a constant and even regulation in time of the status and performance data to the monitoring server and CPU utilization of the monitoring server when the preselected time interval is increased or decreased. 
 
     
     
         4 . The method of  claim 1  further comprising the step:
 dynamically responding to other running instances of the multi-threaded process in the computer network by increasing or decreasing the number of monitoring threads such that each instance of the multi-threaded process will monitor an even share of the available monitored servers. 
 
     
     
         5 . The method of  claim 1  wherein the status and performance data further comprises:
 current alert data, historical alert data, notification data, availability data, status data, performance data, capacity data, security data, command data, schema change data, heartbeat data and internal error data. 
 
     
     
         6 . The method of  claim 1  wherein the status and performance data is provided to the monitoring server as transactional data packets.