Yet another tool to monitor the progress of data through a pipe.

zpv simply copies standard in to standard out and prints progress information to standard error. Like pv, zpv allows a user to see the progress of data through a pipeline. Unlike pv, zpv reports the amount of time spent waiting to read and write to stdin/stdout. This functionality gives more insight into the processes that generate and consume data to help performance profiling and monitoring in long plipelines.

Output to standard error is human readable as well as valid json with one json object per line.

{ "stdin_wait_ms": 26307, "stdout_wait_ms": 1691, "total_time_ms": 27999, "bytes_out": 410473472 }

1. stdin_wait_ms: The amount of time, in milliseconds, that zpv spent waiting for data on standard in.
2. stdout_wait_ms: The amount of time, in milliseconds, that zpv spent waiting for data to be consumed on standard out.
3. total_time_ms: The total amount of time, in milliseconds, that has elapsed.
4. bytes_out: The total number of bytes that have been written to standard out.

1. throughput: bytes_out / total_time_ms
2. overhead created by the zpv tool itself: total_time_ms - ( stdin_wait_ms + stdout_wait_ms )
3. stalled input: stdin_wait_ms continues to increment while bytes_out remains static.
4. stalled output: stdout_wait_ms continues to increment while bytes_out remains static.

zpv has no options.

dd if=/dev/random | zpv >/dev/null