CNS Core: Small: Practical Memory Access Pattern Obfuscation with Algorithm, Application and Architecture Co-designs

CNS 核心：小型：通过算法、应用程序和架构协同设计进行实用内存访问模式混淆

• 批准号: 2152497
• 负责人: Xian-He Sun
• 金额: $ 49.8万
• 依托单位: Illinois Institute of Technology
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-08-01 至 2025-07-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2152497&HistoricalAwards=false
• 关键词: CNS; Core; Small; Practical; Memory

Emerging memory architectures are proposed to address the scalability issues of current computing systems. However, they bring challenges when data security support is needed. An attacker could capture sensitive information in a program through side channels, such as memory access patterns. Moreover, emerging memory systems further complicate security support due to the new properties such as larger sharing space, data persistency, and concurrent accesses. This research aims to enable high-performance and practical memory access pattern obfuscation for current and emerging memory architectures. The approaches include co-designs from the cryptographic algorithms, target applications that need protection, and new memory architectures. The success of this project can keep improving the performance of cryptographic protocols on real computing systems and applications. The outcome will include a set of integrated, cross-domain secure architecture designs and obfuscation schemes. The project outcomes will have significant broader impacts on education, society, and national security. This project will broaden computing participation by recruiting undergraduate researchers and outreach to underrepresented groups. With a closer interaction between security and computer architecture, the proposed research will benefit a broader type of applications and thus help protect national security in the long run.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

新兴的存储器架构被提出来解决当前计算系统的可扩展性问题。然而，当需要数据安全支持时，它们会带来挑战。攻击者可以通过侧通道捕获程序中的敏感信息，例如内存访问模式。此外，新兴的存储器系统进一步复杂化的安全支持，由于新的属性，如更大的共享空间，数据持久化，并发访问。本研究的目的是使高性能和实用的内存访问模式混淆当前和新兴的内存架构。这些方法包括密码算法的协同设计，需要保护的目标应用程序和新的存储器架构。该项目的成功可以不断提高密码协议在真实的计算系统和应用中的性能。其成果将包括一套集成的跨域安全架构设计和混淆方案。该项目的成果将对教育、社会和国家安全产生更广泛的影响。该项目将通过招募本科研究人员和向代表性不足的群体推广来扩大计算参与。随着安全和计算机体系结构之间的更密切的相互作用，拟议的研究将有利于更广泛的应用类型，从而有助于保护国家安全的长期运行。这一奖项反映了NSF的法定使命，并已被认为是值得通过使用基金会的智力价值和更广泛的影响审查标准进行评估的支持。

项目成果

SampleMine: A Framework for Applying Random Sampling to Subgraph Pattern Mining through Loop Perforation

• DOI: 10.1145/3559009.3569658
• 发表时间: 2022-10
• 期刊: Proceedings of the International Conference on Parallel Architectures and Compilation Techniques
• 作者: Peng Jiang;Yihua Wei;Jiya Su;Rujia Wang;Bo Wu

CARE: A Concurrency-Aware Enhanced Lightweight Cache Management Framework

• DOI: 10.1109/hpca56546.2023.10071125
• 发表时间: 2023-02
• 期刊: 2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)
• 作者: Xiaoyang Lu;Rujia Wang;Xian-He Sun

DNNCloak: Secure DNN Models Against Memory Side-channel Based Reverse Engineering Attacks

• DOI: 10.1109/iccd56317.2022.00023
• 发表时间: 2022-10
• 期刊: 2022 IEEE 40th International Conference on Computer Design (ICCD)
• 作者: Yuezhi Che;Rujia Wang