| const EXAMPLES: &str = r#"EXAMPLES: |
| |
| 1) Default Benchmarking with TCP: |
| |
| Start server, perform given benchmark with 10 producers / 10 consumers and |
| 10 streams, sending 1000 message of 10000 byte each per batch, 1000 batches: |
| |
| $ cargo r --bin iggy-bench -- send tcp |
| $ cargo r --bin iggy-bench -- poll tcp |
| $ cargo r --bin iggy-bench -- send-and-poll tcp |
| |
| This will save around 10 GB of data on SSD. |
| Keep in mind that after each benchmark, the server will be NOT be stopped. |
| Server will have to be stopped manually afterwards. |
| |
| 2) Custom Benchmarking with cleanup and visible server stdout: |
| |
| Start server and cleanup after benchmark is done, server stdout will be visible |
| perform TCP benchmark with 5 producers and 5 streams, sending 2000 messages of |
| 100 bytes each per batch, for 1000 batches - 20 GB of data on SSD will be saved: |
| |
| $ cargo r --bin iggy-bench -r -- -c -v send --message-size 2000 --messages-per-batch 1000 --message-batches 1000 --producers 5 --streams 5 tcp |
| |
| |
| 3) Benchmarking with remote server: |
| |
| It's possible to benchmark remote server. In this case, omit the `cleanup` and `verbose` flags, |
| and specify server IP after the transport: |
| |
| $ cargo r --bin iggy-bench -- send --message-size 2000 --messages-per-batch 1000 --message-batches 1000 --producers 5 --streams 5 tcp --server-address 142.250.203.142:8090 |
| |
| 4) Other options: |
| |
| If more options are needed, please refer to the help menu: |
| |
| $ cargo r --bin iggy-bench -r -- --help |
| |
| Each subcommand has it's own help: |
| |
| $ cargo r --bin iggy-bench -r -- send --help |
| $ cargo r --bin iggy-bench -r -- poll --help |
| $ cargo r --bin iggy-bench -r -- send-and-poll --help |
| |
| $ cargo r --bin iggy-bench -r -- send tcp --help |
| $ cargo r --bin iggy-bench -r -- poll tcp --help |
| $ cargo r --bin iggy-bench -r -- send-and-poll tcp --help |
| |
| "#; |
| |
| pub fn print_examples() { |
| print!("{}", EXAMPLES) |
| } |