Do a good corporate website, the server is relatively independent comparative advantage, companies have data stored in a particular metropolitan standalone server, therefore, we should always do the independent server performance monitoring, in order to maintain the security monitoring data . One group is smaller than the following metaphor to give you Analysis. As the disk drive performance analysis are common factors that affect system performance, so managers need to collect performance disk I / O status information to determine optimize overall system performance indicators. Memory utilization and CPU utilization, as managers need to understand a separate server memory utilization, and monitor the current process list occupied by the memory, the threshold set warning when using too high, managers can be known in time. File system capacity service applications require the use of disk space for storage and processing, there is insufficient space may lead to service can not function properly, therefore, managers should always monitor the amount of disk space free. In general, file space usage should not exceed 85% ,, more than once, the administrator should be treated immediately. In the management interface, the user can determine the size of the threshold value based on the corresponding identification. Resource records make a record on a separate server performance resource usage data, and then analyzed to see if the NIC traffic for a period of time have abnormal changes, if the peaks and valleys appear, the need for a separate server to be checked. CPU utilization management personnel need to understand the system corresponding to each CPU utilization, and server monitoring mission is currently occupied by the CPU, ensuring that none of the task to run. Threshold Warning Set CPU, when the CPU load is too large, an alarm can be generated, providing management process to proceed. In addition to performance monitoring and data logging for stand-alone server, but also analyze certain period of time there is no abnormal flow of change, sometimes high and sometimes low if the situation should be done in time to troubleshoot stand-alone server.
