Get Knowledge from Video instead of Content:

Thursday, 24 November 2016

Windows Server Performance Counters

List of all Windows counters used during performance test analysis:

Processor Counters:

1. % Processor Time Counter Indicates the percentage of elapsed time that the processor spends to execute a non-idle thread.
2. % Privileged Time Counter Indicates the percentage of elapsed time that the process threads spent executing code in privileged mode 
3. % Interrupt Time Counter Indicates the time the processor spends receiving and servicing hardware interrupts during sample intervals 
4. Processor Queue Length Counter Indicates the number of threads in the processor queue 
5. Context Switches Counter Indicates the combined rate at which all processors on the computer are switched from one thread to another 
6. System Up Time Counter Indicates the indicator of overall system availability
Windows Server Counters


Memory Counters:

1. Available Bytes Counter Indicates the amount of physical memory available to processes running on the computer 
2. Working Set Counter Indicates the number of resident pages of each process 
3. Pages/sec Counter Indicates the rate at which pages are read from or written to disk to resolve hard page faults
4. Page Reads/sec Counter Indicates that the working set of the process is too large for the physical memory and that it is paging to disk 
5. Pool Nonpaged Bytes Counter Indicates the size of an area of system memory (physical memory used by the operating system) for objects that cannot be written to disk but must remain in physical memory as long as they are allocated 
6. Paged Pool Bytes Counter Indicates memory leaks 
7. Paged Pool Failures Counter Indicates the number of times allocations from the paged pool have failed 
8. Cache Bytes Counter Indicates the size of the static files cache 
9. System Cache Resident Bytes Counter Indicates the number of resident pages allocated to the System File Cache 
10. Committed Bytes Counter Indicates extreme paging leading to slow and irregular response times

I/O Counters:

1. Avg. Disk secs/transfer Counter Indicates physical disk potential bottleneck 
2. % Idle Time Counter Indicates physical disk utilization 
3. Disk Transfers/sec Counter Indicates whether the physical disk is a potential bottleneck 
4. Avg. Disk Queue Length Counter Indicates, although in conjunction with other counters, a potential bottleneck of a disk 
5. Split IO/sec Counter Indicates possible defragmentation  
6. Free Megabytes Counter Indicates logical disk space usage

Network Counters:

1. Bytes Total/sec Counter indicates total throughput 
2. Server Bytes Total/sec Counter indicates overall server utilization in terms of network 
3. Datagrams/sec Counter indicates IP protocol load 
4. Connections Established Counter indicates TCP protocol connection success rate 
5. Segments Received/sec Counter indicates the number of TCP data segments received 
6. % Interrupt Time Counter indicates the time the processor spends on hardware devices interrupts, such as the network card 

Information source: https://www.microfocus.com/media/documentation/loadrunner_and_performance_center_document.pdf


Related Topics:

      4 comments :

      1. Nice list Gagan but if you publish the SLA/Recommended Values for the corresponding counters, that would help more to understand the system.

        ReplyDelete
        Replies
        1. I am not sure if I am completely right.. But these counters will have to be observed as graphs depending on the type of performance tests your run i.e Load Test or Endurance Test .or simply a Performance Regression (Performance Test to compare your latest Deploy,en against baselines) .. In all of these cases the graphs of counters will differ..So there cannot be some industry values defined..

          Delete
      2. Please put some light on the avobe question.

        ReplyDelete