SGXGauge

SGXGauge是由印度理工学院德里分校的研究团队开发的一个综合基准套件，用于评估Intel SGX的性能。该数据集包含10个来自不同领域的真实世界和合成基准，旨在全面评估Intel SGX的关键组件，包括加密/解密、enclave转换和EPC故障等。SGXGauge通过在本地平台和使用基于库的操作系统层（如GrapheneSGX）的平台上的彻底行为特征化，提供了对Intel SGX性能的深入理解。数据集的应用领域包括区块链代码、安全机器学习算法、轻量级Web服务器和安全键值存储等，旨在解决在不可信机器上安全执行应用程序的问题。SGXGauge is a comprehensive benchmark suite developed by a research team from the Indian Institute of Technology Delhi, intended for evaluating the performance of Intel SGX. This suite includes 10 real-world and synthetic benchmarks across diverse domains, aiming to comprehensively assess the key components of Intel SGX, such as encryption/decryption, enclave switching, and EPC faults. SGXGauge provides in-depth insights into Intel SGX performance through thorough behavioral characterization on both local platforms and platforms employing library-based OS layers like GrapheneSGX. Its application fields cover blockchain code, secure machine learning algorithms, lightweight web servers, secure key-value storage, and more, targeting the challenge of securely executing applications on untrusted machines.

搜集汇总

背景与挑战

背景概述

SGXGauge是一个由印度理工学院德里分校开发的综合基准套件，用于评估Intel SGX的性能，包含10个真实世界和合成基准，覆盖加密/解密、enclave转换等关键组件的测试。它通过本地和基于库的平台（如GrapheneSGX）进行行为分析，提供深入性能理解，应用于区块链、安全机器学习、轻量级Web服务器和安全键值存储等领域，旨在解决不可信环境中的安全执行问题。