https://github.com/wunkolo/qcheck
A quick alternative to cksfv for generating and verifying CRC32 checksum files.
https://github.com/wunkolo/qcheck
checksum crc error-detection hash
Last synced: 5 months ago
JSON representation
A quick alternative to cksfv for generating and verifying CRC32 checksum files.
- Host: GitHub
- URL: https://github.com/wunkolo/qcheck
- Owner: Wunkolo
- License: mit
- Created: 2019-12-02T06:22:18.000Z (almost 6 years ago)
- Default Branch: main
- Last Pushed: 2025-02-05T05:15:03.000Z (8 months ago)
- Last Synced: 2025-03-31T05:24:54.276Z (6 months ago)
- Topics: checksum, crc, error-detection, hash
- Language: C++
- Homepage:
- Size: 92.8 KB
- Stars: 19
- Watchers: 3
- Forks: 6
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# qCheck [](LICENSE)
A **q**uick alternative to cksfv for generating and verifying CRC32C checksum files(`.sfv`).
---
qCheck is about **x4** to **x15** times faster than alternatives such as
`rhash` and `cksfv` and reduces algorithmic overhead to being almost
entirely bounded by device IO!
It is recommended to run this program on a fast NVME Solid State Drive to fully
saturate read speed.
---
## Install
### Manual
```
$ git clone https://github.com/Wunkolo/qCheck.git qCheck/src
$ cmake -B qCheck/build -S qCheck/src -DCMAKE_BUILD_TYPE=Release
$ cmake --build qCheck/build --config Release
# install -Dm+x "qCheck/build/qCheck" "/usr/local/bin/qCheck"
```
### Packages
[ArchLinux](https://aur.archlinux.org/packages/qcheck-git)
## Benchmark
Specs:
```
% inxi -IC
CPU: Topology: 10-Core model: Intel Core i9-7900X bits: 64 type: MT MCP L2 cache: 13.8 MiB
Speed: 1201 MHz min/max: 1200/4500 MHz Core speeds (MHz): 1: 1201 2: 1200 3: 1200 4: 1201 5: 1201 6: 1200 7: 1197
8: 1201 9: 1201 10: 1201 11: 1201 12: 1201 13: 1201 14: 1201 15: 1200 16: 1201 17: 1200 18: 1200 19: 1200 20: 1200
Info: Processes: 350 Uptime: 9d 14h 45m Memory: 62.51 GiB used: 19.88 GiB (31.8%) Init: systemd Shell: vim inxi: 3.0.37
% sudo nvme list
Node SN Model Namespace Usage Format FW Rev
---------------- -------------------- ---------------------------------------- --------- -------------------------- ---------------- --------
/dev/nvme1n1 S~~~N~~~9~~~7~~ Samsung SSD 970 EVO 500GB 1 341.05 GB / 500.11 GB 512 B + 0 B 2B2QEXE7
```
Verifying an .sfv file(10GB folder full of +200MB files)
```
% hyperfine --runs 10 -i '../qCheck -c TestCheck.sfv' 'rhash -c TestCheck.sfv' 'cksfv -g TestCheck.sfv'
Benchmark #1: ../qCheck -c TestCheck.sfv
Time (mean ± σ): 1.494 s ± 0.198 s [User: 5.140 s, System: 0.817 s]
Range (min … max): 1.282 s … 1.740 s 10 runs
Benchmark #2: rhash -c TestCheck.sfv
Time (mean ± σ): 7.374 s ± 0.015 s [User: 5.769 s, System: 1.598 s]
Range (min … max): 7.352 s … 7.400 s 10 runs
Benchmark #3: cksfv -g TestCheck.sfv
Time (mean ± σ): 21.376 s ± 0.663 s [User: 19.848 s, System: 1.512 s]
Range (min … max): 21.084 s … 23.260 s 10 runs
Summary
'../qCheck -c TestCheck.sfv' ran
4.94 ± 0.65 times faster than 'rhash -c TestCheck.sfv'
14.31 ± 1.94 times faster than 'cksfv -g TestCheck.sfv'
```
Generating a .sfv file(10GB folder full of +200MB files)
```
% hyperfine --runs 10 "rhash SampleFiles/*" "./qCheck SampleFiles/*" "cksfv SampleFiles/*"
Benchmark #1: rhash SampleFiles/*
Time (mean ± σ): 6.234 s ± 0.145 s [User: 4.798 s, System: 1.424 s]
Range (min … max): 6.133 s … 6.632 s 10 runs
Benchmark #2: ./qCheck SampleFiles/*
Time (mean ± σ): 1.435 s ± 0.105 s [User: 5.146 s, System: 0.610 s]
Range (min … max): 1.289 s … 1.590 s 10 runs
Benchmark #3: cksfv SampleFiles/*
Time (mean ± σ): 20.896 s ± 0.074 s [User: 19.611 s, System: 1.271 s]
Range (min … max): 20.778 s … 20.996 s 10 runs
Summary
'./qCheck SampleFiles/*' ran
4.34 ± 0.33 times faster than 'rhash SampleFiles/*'
14.56 ± 1.07 times faster than 'cksfv SampleFiles/*'
```