LoadRunner Performance Counter Analysis reference value

I. Memory counters

1. pagesinput/sec is the number of pages read from the hard disk to solve the hard error page. The pageinput/sec value can be used to determine a hard error.

Page occurrence rate. Generally, the value is greater than or equal to pagereads/sec.

2. pagereads/sec is used to solve hard errors and read the number of times from the hard disk. If the pagereads/sec ratio persists to 5

Indicates that the memory may be insufficient.

3. Page/sec refers to the number of pages read from, or written to, the disk for parsing hard page errors. Page/sec recommended 00-20 (if the server

There is not enough memory to handle its workload, and this value will remain high. If it is greater than 80, it indicates there is a problem ).

4. Available Mbytes physical memory availability. If the value is small (MB or smaller), it indicates that the memory is insufficient or the program does not

Release memory.

5. cache bytes file system cache. By default, 50% of the available physical memory is used. If the iis5.0 running memory is insufficient, it will

Automatic cache sorting. You need to pay attention to the trend changes of this counter. If you suspect Memory leakage, Please monitor memory \\

Available bytes and memory \ committed bytes to observe memory behavior and monitor what you think may be leaked

Process \ private bytes, Process \ working set, and process \ handle count of the stored process. If

If you suspect that the kernel-mode process causes leakage, you should also monitor memory \ pool nonpaged bytes, memory \\

Pool nonpaged allocs and process (process_name) \ pool nonpaged bytes.

Ii. Service cache counters (Web)

1. filecachehits % is the ratio of File Cache to all cache requirements, reflecting the file cache settings in IIS.

(For a website composed of mostly static webpages) filecachehits % is very good around 80%!

2. filecachehits is the specific value hit by the file cache.

Filecacheflushes indicates the number of File Cache refreshes since the server is started. If the refresh is too slow, the memory will be wasted. If the refresh is too slow

Fast, the objects in the cache will be discarded too frequently and will not play the role of cache. Using filecachehits and filecacheflushes

You can get an appropriate refresh value (refer to setting objectttl, memcachesize, and maxcachefilesize in IIS ).

3. Connection refused: the lower the value, the better. A high value indicates a bottleneck in the network adapter or processor.

Iii. CPU counters

1.% processor time: if the value continuously exceeds 80% or 85%, the bottleneck is the CPU. You can consider adding a processor or changing

A faster processor.

2.% USER time: Indicates CPU-consuming database operations, such as sorting and executing Aggregate functions. If the value is high,

You can consider increasing the index, try to use simple table join, horizontally split large tables, and other methods to reduce this value.

3.% privileged time: (CPU Kernel Time) indicates the percentage of time taken to process the code executed by a thread in privileged mode. If

The parameter value and the "physical disk" parameter value are always high, indicating I/O problems. Consider replacing a faster hard drive system. Additional settings

Tempdb in RAM reduces "Max async Io" and "maxlazy writer Io.

4. processor Queue Length indicates the number of threads in the queue to be processed. This counter displays the processor bottleneck. Queue duration

If the value is greater than 4, the processor may be congested. This counter is the value of a specific time, not the average value of a period of time.

5.% DPC time: the lower the better. In a multi-processor system, if the value is greater than 50% and processor: % Processor
Time is very high. Adding a network adapter may improve performance and the network provided is not saturated.

4. physical disk counters

1.% disk Time %: percentage of the time the selected disk drive was busy providing services for read or write requests. If three counts

So hard disk is not the bottleneck. If only % disk Time is large and both are moderate, the hard disk may be a bottle

Neck. Before recording this counter, run diskperf-YD in the command line window of Windows 2000. If the value exceeds

80%, the memory may leak.

2. avg. Disk queue length: the average number of Read and Write requests (queued for the selected disk in the instance interval. This value should not exceed 1.5 of the number of disks ~ 2
Times. To improve performance, you can add disks. Note: A raid disk actually has multiple disks.
3. Average disk read/write queue length: the average number of read (write) requests (queues.
4. disk reads (writes)/S: Number of disk reads and writes on a physical disk per second. The sum of the two should be smaller than the maximum capacity of the disk device.
5. Average disksec/read: the average time required to read data on this disk in seconds.
6. Average disk SEC/transfer: the average time required to write data to this disk in seconds. The value is the best between 15 and 15 ms ~ Between 30 ms

Is good, 30 ~ 60 ms is acceptable. If 60 ms is exceeded, consider replacing the hard disk or hard disk in Riad mode.
7. bytes total/sec: the transmission and receipt byte rate, including frame characters. Determine the network connection. Whether the speed is a bottleneck. You can use the counter value and

Current network bandwidth comparison

V. Thread counters

1. contextswitches/sec: (instantiate inetinfo and DLLHOST processes) if you decide to increase the size of the thread byte pool, you should monitor these three counts

(Including the above one ). Increasing the number of threads may increase the number of context switches, so that the performance will not increase but decrease. If the context of the ten instances

If the switching value is very high, the size of the thread's byte pool should be reduced.

6. process counters

1. processor/% processor time: Number of processors consumed by the processor. if the service is exclusively for the SQL Server service, the restructuring range cannot exceed 80%.

~ 85%.

2. Page faults/sec: Number of soft page failures per second (including some that can be satisfied directly in the memory and some that need to be read from the hard disk) compared to page/sec only tables

Explicit data cannot be used immediately in a specified work set in the memory.

3. Working set: the number of current bytes in the processing working set. The working set is the memory page recently touched by the thread during processing.

Set. If the available memory on the computer is above the threshold, even if the page is not in use, it will remain in a working set. When available memory drops

To the threshold value, the page will be deleted from the working set. If you need a page, it will return a soft fault to the working set before leaving the main memory.

4. processor/private Bytes: indicates the number of currently allocated bytes that cannot be shared with other processes.

The database is then supplemented.