Software program Performance metrics

If all of the internal technical factors are left behind, we only look at the performance metrics of the application, so generally speaking, the performance of the program can be measured in the following ways.

N Response Time: The time the system responds to user behavior or events. The shorter the response time, the better performance, so we should try to use asynchronous processing in the system design process, so that users can receive receipts as soon as possible, so that the user experience will be better.

N Start-up time: The application system from running to the normal processing of business needs to spend time, for the user, it must be faster start the better, so we should try to use asynchronous loading data in the system design process to start the application, to avoid waiting for all data to be loaded before starting.

N Execution time: a period of code from the beginning to the end of the run, the time used is called the execution time. For execution time, it may not be possible to reduce the time to global, but you can make the user feel that the execution time is shortened by slicing the business logic into multiple contiguous segments of the program.

N Execution Speed: whether the program reacts quickly and whether the response time is short enough. The indicator is associated with the response time and execution time.

N COMPUTE resource allocation: Compute resources, including CPU, memory, disk, and so on, if any one of these allocations is unreasonable, it may cause the whole system to always be in a tight computing resource situation, so the performance impact on the entire system must be devastating.

N Memory allocation: Whether memory allocation is reasonable, excessive memory consumption or presence of leaks, JVM performance also has a certain relationship with memory allocation.

N Disk Throughput: Describes the usage of I/O. IOPS (Input/output per Second) is the number of input/output (or read/write) per second and is one of the key metrics for disk performance. ioPS refers to the number of I/O requests that the system can process per unit of time, and I/O requests are typically read or write data operation requests. For applications where random reads and writes are frequent, such as OLTP (Online Transaction processing), IOPS is a key metric. Another important indicator is data throughput (throughput), which refers to the number of data that can be successfully transferred per unit of time. For applications with a large number of sequential reads and writes, such as VOD (Video OnDemand), more attention is paid to throughput metrics. I/O throughput per second =iopsx average I/O SIZE. As you can see from the formula, the larger the I/O size, the higher the ioPS, the higher the I/O throughput per second. Therefore, we think that the higher the IOPS and throughput values, the better. In fact, for a disk, both parameters have their maximum value, and there is a certain relationship between the two parameters.

N Network throughput: Describes the use of the network. The data in the network consists of one packet, and the firewall consumes resources to deal with each packet. Throughput is the maximum rate that a device can accept without frame loss. The test method is to send a certain number of frames at a certain rate in the test, and calculate the frames transmitted by the device to be measured, and if the sending frame is equal to the number of frames received, the sending rate is increased and re-tested, if the received frame is less than the sent frame, the sending rate is re-tested until the final result is reached The throughput test results are expressed as "bits per second" or "bytes per second".

N Load carrying Capacity: When the system pressure rises, the system execution speed, response time rise curve is flat. Load endurance is associated with many factors such as computing resources, memory, disk, and network.

Please pay attention to Uncle Mike every 10 o'clock in the evening said, let us communicate and study together.