LoadRunner performance indicator Positioning System Bottleneck

Determine CPU bottleneck

1,% Processor timeAverage value greater than 95

2,Processor Queue LengthGreater than 2 (greater than the number of processors + 1). You can determine the CPU bottleneck

3. The CPU idle time is zero (Zero percent idle CPU)

4. High CPU usage time(% USER time)

5. Excessive CPU usage(% Priviliaged time: More than 90% or 95% in the long term)

Note:

% USER time (processor_total) indicates CPU-consuming database operations, such as sorting and executing Aggregate functions. If the value is very high, you can consider increasing the index and try to reduce the value by using simple table join and horizontal table segmentation methods.

If processor queue length is found to display more than 2 queues, but the processor utilization is always low, it may be better to solve the problem of processor congestion. Here, the processor is generally not a bottleneck.

 

Determine memory bottleneck and Memory leakage

1. If memory leakage occurs,Process \ private bytesCounter andProcess \ Working SetThe counter value is usually increasedAvaiable bytes.

2. IfAvailable MbytesIf the value (number of remaining physical memory) is small (4 MB or smaller), the total memory on the computer may be insufficient or a program does not release the memory.

 

Locate disk bottlenecks

1,% Disk TimeAndAVG. Disk Queue LengthThe value (should not exceed 1.5 to 2 times of the number of main axes that constitute the physical disk) is very high, while the page reads/sec page reading speed is very low, there may be disk bottle diameter.

2,Physical disk \ disk reads/sec and disk writes/secDisk bottlenecks are possible if the MS is greater than 20

3,AVG. Disk SEC/TransferThe average time of data written to the disk, in seconds. Generally, it is best to define a value smaller than 15 ms. It is good between 15-30 ms and acceptable between 30-60 ms, if the disk size exceeds 60 ms, you need to consider replacing the hard disk or hard disk raid mode.

4,Disk transfers/secThe read/write operation rate on this disk. Normal value <(Disk Bytes/sec)/3. This value is too large, indicating that the IO speed required by the system is close to the maximum speed of the hard disk. You need to replace the hard disk faster.

Note: If a RAID device is used, the % disk Time counter indicates a value greater than 100%.

 

Locate Network bottleneck Byte total/secIndicates the speed at which bytes are received and sent in the network. You can use this counter to determine whether the network has a bottleneck (reference value: This counter is different from the network bandwidth, <50%)

 

SQL Server Problems

1,Sqlserver: cachemanage-> cache hit ratioThe hit rate of data found in the cache. If the value is less than 85%, the memory is faulty.

2,Sqlserver: locks-> lock waits/secDisplays the number of lock requests per second that force other processes to wait before the current process is completed. If the value is always greater than 0, the transaction is faulty.

3,Sqlserver: databases-> transactions/secNumber of transactions started for the database per second

 

From: http://www.cnblogs.com/waitrabbit/archive/2010/04/15/1712787.html

LoadRunner performance indicator Positioning System Bottleneck