mirror of
https://github.com/NotAShelf/tct.git
synced 2025-10-05 08:22:10 +00:00
docs: update README for new cli handling options
Some checks failed
Build & Test CI / build (push) Has been cancelled
Some checks failed
Build & Test CI / build (push) Has been cancelled
This commit is contained in:
parent
4f97aa3d83
commit
c6413a0e58
1 changed files with 46 additions and 35 deletions
81
README.md
81
README.md
|
@ -1,8 +1,8 @@
|
||||||
# tct
|
# tct
|
||||||
|
|
||||||
**tct** (**t**cp **c**onnection **t**imer) is a quick and minimal program (< 70
|
**t**cp **c**onnection **t**imer (tct) is a miniscule utility to help determinee
|
||||||
LoC!) that helps determine the "optimal" number of parallel TCP requests for
|
the "optimal" number of parallel TCP requests for your network connection, for a
|
||||||
your network connection.
|
given target.
|
||||||
|
|
||||||
It performs a series of tests (following your desired configuration) by
|
It performs a series of tests (following your desired configuration) by
|
||||||
incrementally increasing the number of parallel requests and measuring the time
|
incrementally increasing the number of parallel requests and measuring the time
|
||||||
|
@ -11,6 +11,48 @@ taken for each test.
|
||||||
The optimal number is identified as the point where adding more parallel
|
The optimal number is identified as the point where adding more parallel
|
||||||
requests does not significantly reduce the overall time taken.
|
requests does not significantly reduce the overall time taken.
|
||||||
|
|
||||||
|
## Usage
|
||||||
|
|
||||||
|
```bash
|
||||||
|
Usage:
|
||||||
|
tct [flags]
|
||||||
|
|
||||||
|
Flags:
|
||||||
|
-d, --delay duration Delay between requests
|
||||||
|
-h, --help help for tct
|
||||||
|
-m, --max int Maximum number of parallel requests (default 100)
|
||||||
|
-u, --url string URL to fetch (default "http://example.com")
|
||||||
|
-v, --version version for tct
|
||||||
|
```
|
||||||
|
|
||||||
|
For example:
|
||||||
|
|
||||||
|
```bash
|
||||||
|
tct --max 200 --delay 100ms --url "http://yourtargeturl.com"
|
||||||
|
```
|
||||||
|
|
||||||
|
Replace `"http://yourtargeturl.com"` with the actual URL you wish to test
|
||||||
|
against. You can also omit the URL and use an IP address instead. For example,
|
||||||
|
`8.8.8.8` for Google or `1.1.1.1` for Cloudflare. You may notice differences
|
||||||
|
between target URLs as a result of different distances to different hosts, or
|
||||||
|
different network setups.
|
||||||
|
|
||||||
|
The `--max` parameter specifies the maximum number of parallel requests to test,
|
||||||
|
and `--delay` sets the interval between each request.
|
||||||
|
|
||||||
|
> [!TIP]
|
||||||
|
> You are strongly advised to use the delay option. I have observed high latency
|
||||||
|
> while running with the default 0 second delay, which is likely some form of
|
||||||
|
> throttling by the host. If you test against an URL that you _know_ does not
|
||||||
|
> throttle connections, then you may consider omitting `-delay`.
|
||||||
|
|
||||||
|
### Flags
|
||||||
|
|
||||||
|
- `--url`: The URL to fetch.
|
||||||
|
- `--max`: Maximum number of parallel requests to test. Default is `100`
|
||||||
|
- `--delay`: Delay between requests. Can be specified as a duration (e.g.,
|
||||||
|
`500ms`). Default is `0` (i.e. no delay)
|
||||||
|
|
||||||
## Motivation
|
## Motivation
|
||||||
|
|
||||||
The [Nix Package Manager](https://github.com/NixOS/nix) has an option called
|
The [Nix Package Manager](https://github.com/NixOS/nix) has an option called
|
||||||
|
@ -25,37 +67,6 @@ set `http-connections` in.
|
||||||
Do keep in mind that this is not 100% accurate. There are many factors that may
|
Do keep in mind that this is not 100% accurate. There are many factors that may
|
||||||
affect the results of network related tests.
|
affect the results of network related tests.
|
||||||
|
|
||||||
## Usage
|
|
||||||
|
|
||||||
```bash
|
|
||||||
tct -url="http://yourtargeturl.com" -max=200 -delay=500ms
|
|
||||||
```
|
|
||||||
|
|
||||||
Replace `"http://yourtargeturl.com"` with the actual URL you wish to test
|
|
||||||
against. You can also omit the URL and use an IP address instead, for example,
|
|
||||||
`8.8.8.8` for Google or `1.1.1.1` for Cloudflare. You may notice differences
|
|
||||||
between target URLs.
|
|
||||||
|
|
||||||
The `-max` parameter specifies the maximum number of parallel requests to test,
|
|
||||||
and `-delay` sets the interval between each request.
|
|
||||||
|
|
||||||
<!-- deno-fmt-ignore-start -->
|
|
||||||
|
|
||||||
> [!NOTE]
|
|
||||||
> You are strongly advised to use the delay option. I have observed high
|
|
||||||
> latency while running with the default 0 second delay, which is likely some
|
|
||||||
> form of throttling by the host. If you test against an URL that you _know_
|
|
||||||
> does not throttle connections, then you may consider omitting `-delay`.
|
|
||||||
|
|
||||||
<!-- deno-fmt-ignore-end -->
|
|
||||||
|
|
||||||
## Flags
|
|
||||||
|
|
||||||
- `-url`: The URL to fetch.
|
|
||||||
- `-max`: Maximum number of parallel requests to test. Default is `100`
|
|
||||||
- `-delay`: Delay between requests. Can be specified as a duration (e.g.,
|
|
||||||
`500ms`). Default is `0` (i.e. no delay)
|
|
||||||
|
|
||||||
## Notes
|
## Notes
|
||||||
|
|
||||||
- Remember to adjust the `-max` and `-delay` parameters based on your network
|
- Remember to adjust the `-max` and `-delay` parameters based on your network
|
||||||
|
@ -63,7 +74,7 @@ and `-delay` sets the interval between each request.
|
||||||
any assumptions, but the default values will not be suitable for all testing
|
any assumptions, but the default values will not be suitable for all testing
|
||||||
conditions.
|
conditions.
|
||||||
- tct will try to always exit gracefully when you, e.g., kill the program with
|
- tct will try to always exit gracefully when you, e.g., kill the program with
|
||||||
<kbd>ctrl+c</kbd>.
|
<kbd>Ctrl+C</kbd>.
|
||||||
|
|
||||||
## Contributing
|
## Contributing
|
||||||
|
|
||||||
|
|
Loading…
Add table
Add a link
Reference in a new issue