Peer-to-Peer Netw. Appl. DOI 10.1007/s12083-016-0481-0 Protecting lightweight block cipher implementation in mobile big data computing A GPU-based approach Weidong Qiu 1 · Bozhong Liu 1 · Can Ge 1 · Lingzhi Xu 1 · Xiaoming Tang 1 · Guozhen Liu 1 Received: 13 April 2016 / Accepted: 8 July 2016 © Springer Science+Business Media New York 2016 Abstract The Mobile Big Data Computing is a new evo- lution of computing technology in data communication and processing. The data generated from mobile devices can be used for optimization and personalization of mobile services and other profitable businesses. Mobile devices are usually with limited computing resources, thus the security mea- sures are constrained. To solve this problem, lightweight block ciphers are usually adopted. However, due to the eas- ily exposed environment, lightweight block ciphers are apt to suffer from differential power attack. To counteract this attack, Nikova et al. proposed a provably secure method, namely sharing, to protect the cipher’s implementation. But the complexity of sharing method is so high, making this method not practical. To address this issue, in this paper, we propose a GPU-based approach of sharing a 4-bit S-box by automatic search. GPU is a promising acceleration hard- ware with powerful parallel computing. By analyzing the sharing method carefully, we devise an optimal approach, namely OptImp, that improves the performance massively. The experiment results show that the proposed approach can achieve up to 300 times faster than the original method. With our approach, the sharing method can be used to protect lightweight block ciphers in practice. Keywords Mobile big data · Lightweight block cipher · Threshold implementation · GPU optimization  Weidong Qiu qiuwd@sjtu.edu.cn Bozhong Liu liu.bo.zhong@gmail.com 1 School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China 1 Introduction The arrival of Big Data era, the proliferation of wireless technology and the development of mobile technology have led to the Mobile Big Data Computing, a new evolution of computing technology in data communication and process- ing [24]. Mobile computing enables users to access various types of services anywhere and anytime using lightweight mobile devices (e.g., tablets, smart phones) [23, 25]. Mobile is particularly well-suited to a big data scenario, because even when we are ostensibly not using our phones, a large quantity of data are still being created, such as the feeds, the Push notifications and other App services. These data can be used for the optimization and personalization of mobile ser- vices, the real-time hyper-local advertising, as well as other profitable businesses [14]. Mobile devices are usually with limited computing resources. As the advance of mobile technology, the new generation devices are getting smaller with limited envi- ronment and power consumption. Because the implementa- tion of the security is bounded to the available computing resources, the small devices are not able to offer comprehen- sive and robust security measures. As a result, lightweight block cipher is usually used to protect users’ sensitive infor- mation in mobile big data [5]. It is a set of well-designed cryptographic algorithms that are suitable for the limited environment and low-cost products [4]. However, due to the limited and easily exposed environ- ment, lightweight block ciphers on the mobile devices are apt to suffer from side-channel attack [9, 22], especially dif- ferential power analysis (DPA) attack [10]. DPA aims to access the secret key by analyzing the leaked information from the intermediate result. It has been shown to be a powerful tool to break many lightweight block ciphers [18, 20]. In order to counteract DPA attacks, several different