An Efficient Profiling-Based Side-Channel Attack on Graphics Processing Units

The encryption/decryption algorithms have been ported to GPU platforms to take advantage of the GPUs’ high-throughput computing capability. The downside of moving the cryptographic algorithms onto GPUs, however, is that the vulnerability of side-channel attacks for GPUs has not been well studied and the confidential information may be under a great risk by processing encryption on GPUs. In this paper, we proposed to leverage a profiling-based side-channel attack (SCA) to expose GPUs’ side-channel vulnerability and the weakness of security services provided by GPUs. Our results show that GPUs are particularly vulnerable to profiling-based side-channel attacks and need to be protected against side-channel threats. Especially, for AES-128, the proposed method can recover all key bytes in less than 1 min, outperforming all prior SCAs we know.

[1]  Yunsi Fei,et al.  A novel cache bank timing attack , 2017, 2017 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).

[2]  David R. Kaeli,et al.  A complete key recovery timing attack on a GPU , 2016, 2016 IEEE International Symposium on High Performance Computer Architecture (HPCA).

[3]  Nael B. Abu-Ghazaleh,et al.  Rendered Insecure: GPU Side Channel Attacks are Practical , 2018, CCS.

[4]  Danfeng Zhang,et al.  RCoal: Mitigating GPU Timing Attack via Subwarp-Based Randomized Coalescing Techniques , 2018, 2018 IEEE International Symposium on High Performance Computer Architecture (HPCA).

[5]  David R. Kaeli,et al.  A Novel Side-Channel Timing Attack on GPUs , 2017, ACM Great Lakes Symposium on VLSI.

[6]  David R. Kaeli,et al.  Power Analysis Attack of an AES GPU Implementation , 2018, J. Hardw. Syst. Secur..

[7]  Ankang Zhang,et al.  Parallel AES algorithm for fast Data Encryption on GPU , 2010, 2010 2nd International Conference on Computer Engineering and Technology.

[8]  Keshab K. Parhi,et al.  GPU accelerated elliptic curve cryptography in GF(2m) , 2010, 2010 53rd IEEE International Midwest Symposium on Circuits and Systems.

[9]  Takakazu Kurokawa,et al.  AES Encryption Implementation on CUDA GPU and Its Analysis , 2010, 2010 First International Conference on Networking and Computing.

[10]  Giovanni Agosta,et al.  Design of a parallel AES for graphics hardware using the CUDA framework , 2009, 2009 IEEE International Symposium on Parallel & Distributed Processing.

[11]  S.A. Manavski,et al.  CUDA Compatible GPU as an Efficient Hardware Accelerator for AES Cryptography , 2007, 2007 IEEE International Conference on Signal Processing and Communications.

[12]  Yiwen Gao,et al.  Cache-Collision Attacks on GPU-Based AES Implementation with Electro-Magnetic Leakages , 2018, 2018 17th IEEE International Conference On Trust, Security And Privacy In Computing And Communications/ 12th IEEE International Conference On Big Data Science And Engineering (TrustCom/BigDataSE).

[13]  David R. Kaeli,et al.  Side-channel power analysis of a GPU AES implementation , 2015, 2015 33rd IEEE International Conference on Computer Design (ICCD).