摘要
文中在GPU硬件解密加速技术领域中,提出了对规则化口令穷举时,口令在设备端自动生成的技术。采用设备端每个并行线程唯一具有的线程号,建立线程号与口令的对应规则,以规则代替IO操作,并基于其中的一种对应规则进行了具体实现。最后以一个约20MB的规则口令库,对GPU口令自生成技术在暴力破解中的应用性能进行了测试。
The paper describes a new method for improving GPU hardware decryption efficiency through ruly password＇s self- manufacturing at device point. A series of well-devised models based on thread unique ID are adopted, which establish the relationship between thread ID and password in replacing IO operation and implement this improvement. Furthermore, one of these models is particularly realized, and the test of 20 M password collection clearly indicates this great improvement.
出处
《信息安全与通信保密》
2013年第3期82-84,共3页
Information Security and Communications Privacy
关键词
GPU优化技术
规则口令
自生成
GPU acceleration
ruly password
self-manufacturing
