Common performance indicators
The number of transactions completed during a fixed time interval of "throughput". Typically, the number of requests processed in 1 seconds, in units: Transactions per second (TPS).
Average throughput over a period of time. Cannot reflect instantaneous changes in throughput.
Peak throughput The maximum amount of throughput over time. is one of the important indicators used to assess system capacity.
The minimum throughput value for a period of time. If the minimum value is close to 0, the system has a "card" phenomenon.
The 70% throughput concentration interval calculates 70% of the throughput concentration interval by counting the throughput boundary values of 15% and 85%. The more concentrated the interval, the more stable the throughput.
The processing time of the transaction for the response time. Usually refers to the time interval, in milliseconds, that is emitted from a request, returned to the server for processing, and then received to complete the response data.
Average response time over a period of time. Response time fluctuations are not reflected.
"Intermediate response time" is the median response time for a period of time, 50% response time, and half of the server response time is below that value and the other half is above that value.
"90% response Time" 90% of transaction response time is smaller than this value over time. Response to the overall response speed, and the 10% timeout rate above that value. is one of the important indicators used to assess system capacity.
Minimum response time for the Min. reflects the fastest processing power of the service.
Maximum response time maximum. Reflects the server's slowest processing power.
"CPU utilization" 1-CPU idle rate, which indicates CPU usage, reflects the utilization of system resources.
Server system resources: Native CPU utilization, memory consumption disk read-write metrics
Network usage: Base throughput rate
Transaction speed: such as average logon time, operation average response time
What are the common performance metrics in server performance testing?