在多FPGA上的高性能VPN口令恢复方法

High-performance VPN Password Recovery Method on Multiple FPGAs

随着国家不断加大对网络安全领域的重视程度,工信部出台规定私人不得自行建立和租用VPN.但由于VPN通信的保密性导致有关部门难以有效监管.因此,实现一种高效的VPN口令恢复方案具有重要意义.通过对VPN使用的身份验证协议以及数据加密协议的分析,本文提出了一种针对使用PPTP隧道协议进行VPN通信的破解模式—VPN-DES模式,并实现了基于多组全流水线架构的FPGA口令恢复方案. VPN-DES模式能将需要暴力穷举的口令空间缩小到2^(56).使用VPN-DES模式恢复的口令不仅可以实现身份认证,还可以用于破解VPN加密的通信数据.采用全流水线架构的FPGA设计在充分发挥FPGA高性能计算的优势的同时,还能在有限时间内遍历全口令空间,因此本方案拥有100%的破解率.实验结果表明,与采用CPU、GPU暴力破解相比,该方案性能比GPU提高12. 8倍,相比CPU则提高5万倍左右.

As the country pays more attention on the field of cyber security,the government promulgated regulations that individuals must not establish and lease VPNs on their ow n. However,due to the confidentiality of VPN communications,it is difficult for the relevant authorities to effectively supervise them. Therefore,to achieve an efficient VPN password recovery program is of great significance. By analyzing the authentication protocol and data encryption protocol used by VPN,this paper proposes a VPN-DES mode for VPN communication using PPTP tunneling protocol and implements FPGA password recovery based on multi-group full pipelined architecture Program. VPN-DES mode can reduce the need for brute force password space to 256. Password recovery using VPN-DES mode not only enables authentication,but also can be used to crack VPN-encrypted communication data. FPGA Design with Full-Pipe Architecture not only fully exploiting the advantages of FPGA high-performance computing,but also can traverse full password space for a limited time,so this solution has a 100% cracking rate. The experimental results show that compared with the use of CPU,GPU brute force solution,the performance of the program is 12. 8 times higher than the GPU,compared with the CPU is increased by 50,000 times.