A bottleneck occurs when performance is restricted by a single resource (i.e. CPU, Memory or Disk I/O) â€“ and the resource is limited by its capacity to deliver. This will occur frequently on a stressed computer system â€“ when the system is unable to respond at an acceptable level. On a fully loaded system, the elimination of the first Bottleneck will result in a further Bottleneck when another resource in turn becomes overloaded. There is always another Bottleneck!
The Current speed of the processor in MHz.
The number of reads of data from the disk.
The number of writes of data to the disk.
Disk Queue Length
Average number of read or write requests that were queued for the selected disk during the sample interval.
IO Operations Read/Write
Rate at which the process is issuing I/O operations. This counts all I/O activity generated by the process, including file, network, and device I/O operations.
The total Physical Memory is shown, as well the available Physical Memory. Under heavy loads, the available Physical Memory change for every scan.
Memory: Soft Faults
The Number of faults which occur when a page is not found in the file system cache and must be retrieved from elsewhere in memory.
The total Virtual Memory is shown, as well the available Virtual Memory. Make sure that you have sufficient Virtual Memory configured.
The number of logical processors for the current instance of the processor. With say 4 processors, if one of them is 100% busy, the CPU shown will be 25%.
There are no magic numbers in tuning a System. For example, 100% CPU utilisation will be good when running a solitary job for the fastest completion time – and bad for interactive users. Rather than looking for a Magic Number, concentrate on reducing the height of the peaks in the Charts.
Measurement is the Key
Performance optimisation is not a mystery. If you want to optimise your system, you need to know what is happening and you need to measure what is happening.
Packets Per Sec
Rate at which packets are sent and received on the network interface.
A page fault occurs when a process requires code or data that is not in memory. Faults that require a disk accesses can cause response delays.
This involves identifying the many causes of system degradation: Bottlenecks, cleaning out non-essential files, defragmenting the hard drives, removing unwanted programs that make start-up time slow, stopping unnecessary services that don't need to run. And lots more!
Elapsed time that all of the threads of a process used the processor to execute instructions.
The time in Milliseconds to extract measurement information in scanning the Windows system. This is a invaluable indicator of system health. It aggregates all system activity.
Use the "Programs and Features" routine to see what programs have been installed on your system. Uninstall any software no longer in use. Files relating to the uninstalled program frequently remain – search for the uninstalled Program and delete the folder and contents.
It is important to check for high CPU usage. It may be a sign of a malicious process. Even genuine Windows Services can exhibit abnormal behaviour. Make sure that each Service is really needed. Use a reliable anti-virus checker to ensure that your system has no nasties.
Windows Management Instrumentation (WMI) provides an extensive interface to the activity of the Windows Operating System.