[Linux Performance Testing Tool] VMSTAT

VMSTAT

NAME: Vmstat: Reports virtual memory statistics Syntax: Vmstat [-a] [-n] [-t] [-S unit] [delay [count] vmstat [-s] [-n] [-S unit] vmstat [-m] [-n] [delay [count] vmstat [-d] [-n] [delay [count] vmstat [-p disk partition] [-n] [delay [count] vmstat [-f] vmstat [-V] Description: Vmstat reports processor, memory, page, block io, traps, and cpu activity information. The first report is the average value after the service is started. Other reports only delay the sampling within [delay. However, the memory and processor reports are a snapshot. Option:-A switches between active and inactive memory.-f indicates that the number of forks after the system is started contains the same number of fork, vfork, and clone system cals and task created. Each thread-based process represents one or more tasks. This value cannot be reproduced. -T add a timestamp to the output-m Displays slabinfo, (I don't know what it is)-n the header will only be called once during the call period (by default, it is only called once) -s: display the time Count and memory statistics, and display the non-reproducible Delay sampling interval Count sampling times-d: display each disk (device) -w: expands columns with large memory (no difference from the default value in the 5 GB test environment) -p: statistical information of each partition.-S: K or M.-V: View version information. Report: VM Mode (default mode)Procs r: Process waiting for execution B: Number of blocked processes (non-disruptive sleep) Memory Swpd: virtual Memory usage Free: available Memory Buff: Memory Cache used for buffer: memory Inact used for cache: inactive memory Active: Active memory Swap Si: Memory read from swap per second So: Number of writes to swao per second IO Bi: number of blocks read from Block devices per second Bo: number of blocks written to block devices per second System In: number of interruptions per second In clock Cs: Number of context switches per second CPU: Us: user Mode usage Sy: System Mode usage Id: idle rate Wa: Waiting For IO rate, in id St: stolen rate DISK mode (-d) Reads Total: total number of read completions Merged: Grouped reads (multiple reads, Merged to one) Sectors: number of successfully read Sectors MS: The number of milliseconds consumed by reading Writes is similar to reads. I don't explain IO Cur: I/O in progress (I don't know what it means) S: time partition mode spent on IO (-p) reads: Number of Reads on the partition Read sectors: Number of Read sectors Writes: Number of writes on the partition Requested Writes: Total number of write requests.