Research on Key-Value Data Storage and Operation Optimization Method Based on Confidential Computing

doi:10.3969/j.issn.1671-1122.2025.11.005

Abstract

Abstract:

Confidential computing aims to protect data in use by making it inaccessible to applications in main memory through a hardware-based TEE. To optimize the paging overhead and system I/O performance of confidential computing technology during operation, this article implemented a key-value data storage and operation optimization method based on confidential computing in light of SGX1 technology for key-value data query and update operations. This approach stored frequently queried key-value data in plaintext within the TEE’s min-heap, while the primary data structures were maintained as ciphertext hash tables in unprotected main memory. By leveraging the TEE to optimize hot data queries, the method also enhanced cold data queries and updates in main memory through the use of data tags. Experimental results demonstrate that this method is more efficient than directly encrypting and decrypting key-value data for query and update operations, effectively reducing paging overhead and improving system I/O performance.

Key words: confidential computing, SGX, key-value data storage

CLC Number:

TP309

REN Fengyuan, ZHANG Xinyue, LEI Zhuoyi, LIU Xin. Research on Key-Value Data Storage and Operation Optimization Method Based on Confidential Computing[J]. Netinfo Security, 2025, 25(11): 1707-1717.

Figures/Tables 15

References 28

[1]	ZHANG Fengwei, ZHOU Lei, ZHANG Yiming, et al. Trusted Execution Environment: Status and Prospects[J]. Journal of Computer Research and Development, 2024, 61(1): 243-260.
	张锋巍, 周雷, 张一鸣, 等. 可信执行环境:现状与展望[J]. 计算机研究与发展, 2024, 61(1):243-260.
[2]	SHANG Ketong, LU Fang, HUANG Ke, et al. Cluster Nodes Integrity Attestation and Monitoring Scheme for Confidential Computing Platform[C]// IEEE. The 22nd International Conference on Trust, Security and Privacy in Computing and Communications(TrustCom). New York: IEEE, 2023: 740-749.
[3]	XU Tao, KONG Shuaidi, LIU Caihua, et al. Overview of Heterogeneous Confidential Computing[J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(3):755-770.
	徐涛, 孔帅迪, 刘才华, 等. 异构机密计算综述[J]. 吉林大学学报(工学版), 2025, 55(3):755-770.
[4]	LI Zhiang, MASHIMA D, ONG W S, et al. On Practicality of Using ARM TrustZone Trusted Execution Environment for Securing Programmable Logic Controllers[C]// ACM. The 19th ACM Asia Conference on Computer and Communications Security. New York: ACM, 2024: 947-961.
[5]	SCHUNTER M. Intel Software Guard Extensions: Introduction and Open Research Challenges[EB/OL]. (2016-10-28)[2025-07-10]. https://doi.org/10.1145/2995306.299530.
[6]	COSTAN V, DEVADAS S. Intel SGX Explained[D]. Cambridge: Massachusetts Institute of Technology, 2016.
[7]	LI Wei, FENG Wei, QIN Yu, et al. Scheme of Runtime Monitoring in Confidential Computing with Dynamic Integrity Measurement[J]. Journal of Computer Research and Development, 2024, 61(10): 2482-2500.
	李为, 冯伟, 秦宇, 等. 基于动态完整性度量的机密计算运行时监控方案[J]. 计算机研究与发展, 2024, 61(10):2482-2500.
[8]	FENG Erhu, LU Xu, DU Dong, et al. Scalable Memory Protection in the PENGLAI Enclave[C]// USENIX. The 15th USENIX Symposium on Operating Systems Design and Implementation(OSDI 21). Berkely: USENIX, 2021: 275-294.
[9]	YANG Zuoru, LI Jingwei, LEE P P. Secure and Lightweight Deduplicated Storage via Shielded Deduplication-Before-Encryption[C]// USENIX. 2022 USENIX Annual Technical Conference(USENIX ATC 22). Berkeley: USENIX, 2022: 37-52.
[10]	HA Guanxiong, GE Xiaowei, JIA Chunfu, et al. Revisiting SGX-Based Encrypted Deduplication via PoW-Before-Encryption and Eliminating Redundant Computations[J]. IEEE Transactions on Dependable and Secure Computing, 2025, 22(3): 2037-2053. doi: 10.1109/TDSC.2024.3476288 URL
[11]	LUTSCH A, EL-HINDI M, HEINRICH M, et al. Benchmarking Analytical Query Processing in Intel SGXv2[C]// OpenProceedings. The 28th International Conference on Extending Database Technology(EDBT 2025). Barcelona: OpenProceedings, 2025: 516-528.
[12]	PANDITH O S. SGXFault: An Efficient Page Fault Handling Mechanism for SGX Enclaves[J]. IEEE Transactions on Dependable and Secure Computing, 2024, 21(3): 1173-1178. doi: 10.1109/TDSC.2023.3268169 URL
[13]	PARK J, KANG S, LEE S, et al. Hardware-Hardened Sandbox Enclaves for Trusted Serverless Computing[J]. ACM Transactions on Architecture and Code Optimization, 2024, 21(1): 1-25.
[14]	ASLANTAS B, PEKTAS E, BAYDERE S. Distributed Key Value Store for IoT Edge Devices[C]// IEEE. The 9th International Conference on Computer Science and Engineering(UBMK). New York: IEEE, 2024: 1-6.
[15]	CHEN Shanshan, MA Zhenyu, GAO Jun, et al. HSKV: A Key-Value Storage Scheme Using Hash to Reduce Write Pause[C]// IEEE. 3rd International Conference on Frontiers Technology of Information and Computer(ICFTIC). New York: IEEE, 2021: 453-456.
[16]	RAHMAN M H, AL ABID F B, ZAMAN M N, et al. Optimizing and Enhancing Performance of Database Engine Using Data Clustering Technique[C]// IEEE. 2015 International Conference on Advances in Electrical Engineering(ICAEE). New York: IEEE, 2015: 198-201.
[17]	LI Dewen, HUANG Wenjun, LI Long. Design of Real-Time Database for Industrial Control System[C]// IEEE. The 10th IEEE International Conference on Control and Automation(ICCA). New York: IEEE, 2013: 12-14.
[18]	MESSAOUD A A, MOKHTAR S B, SIMONET-BOULOGNE A. TruShare: Confidential Key-Value Store for Untrusted Environments[C]// IEEE. The 20th IEEE European Dependable Computing Conference(EDCC). New York: IEEE, 2025: 1-9.
[19]	SHEN Chenkai, FAN Lei. PLDB: Protecting LSM-Based Key-Value Store Using Trusted Execution Environment[C]// IEEE. The 22nd International Conference on Trust, Security and Privacy in Computing and Communications(TrustCom). New York: IEEE, 2023: 762-771.
[20]	LU Jinkang, XIE Pping. HotKey-Cache: A HotKey-Separate Cache to Improve Performance for Key-Value Store[C]// IEEE. The 2nd IEEE International Conference on Cloud Computing, Big Data Applications and Software Engineering(CBASE). New York: IEEE, 2023: 286-290.
[21]	NAM H, PARK J, LEE S, et al. Hints for Memory Optimization of Asylo-Based SGX Enclave Applications[C]// IEEE. 2020 International Conference on Information and Communication Technology Convergence(ICTC). New York: IEEE, 2020: 1685-1690.
[22]	LEE J, SON S, BAE J, et al. SkipLSM: Fast Retrieval of Hot Key-Value Pairs on LSM Tree[C]// ACM. The 40th ACM/SIGAPP Symposium on Applied Computing. New York: ACM, 2025: 396-405.
[23]	KOO J, BAE J, YUK M, et al. All-Flash Array Key-Value Cache for Large Objects[C]// ACM. The 18th European Conference on Computer Systems. New York: ACM, 2023: 784-799.
[24]	JOHNSON S, MAKARAM R, SANTONI A, et al. Supporting Intel^® SGX on Multi-Socket Platforms[R]. Santa Clara: Intel Corporation, 843058, 2021.
[25]	HUANG Jianming, YU Hua. Root Crash Consistency of SGX-Style Integrity Trees in Secure Non-Volatile Memory Systems[C]// IEEE. 2023 IEEE International Symposium on High-Performance Computer Architecture(HPCA). New York: IEEE, 2023: 152-164.
[26]	LI Mingyu, XIA Yubin, CHEN Haibo. Memory Optimization System for SGXv2 Trusted Execution Environment[J]. International Journal of Software and Informatics, 2022, 12(3): 285-307. doi: 10.21655/ijsi.1673-7288.00287 URL
[27]	TRAMÈR F, BONEH D. Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardware[EB/OL]. (2019-02-27)[2025-07-10]. https://doi.org/10.48550/arXiv.1806.03287.
[28]	KIM T, PARK J, WOO J, et al. ShieldStore: Shielded In-Memory Key-Value Storage with SGX[C]// ACM. The 14th EuroSys Conference 2019. New York: ACM, 2019: 1-15.

查询操作比例	正常查询吞吐量/ （指令数·s^-1）	优化查询1吞吐量/ （指令数·s^-1）	优化查询2吞吐量/ （指令数·s^-1）	性能提升1	性能提升2
100%	3995	14283	14327	257.52%	258.62%
90%	3509	8472	10955	141.44%	212.20%
50%	2635	3337	5660	26.64%	114.80%

查询操作比例	正常查询吞吐量/ （指令数·s^-1）	优化查询1吞吐量/ （指令数·s^-1）	优化查询2吞吐量/ （指令数·s^-1）	性能提升1	性能提升2
100%	1014	3812	8047	275.94%	693.59%
90%	1100	3494	7088	217.64%	544.36%
50%	1330	2474	4820	86.02%	262.41%

查询操作比例	正常查询吞吐量/ （指令数·s^-1）	优化查询1吞吐量/ （指令数·s^-1）	优化查询2吞吐量/ （指令数·s^-1）	性能提升1	性能提升2
100%	945	3943	8159	317.25%	763.39%
90%	1092	3579	7160	227.75%	555.68%
50%	1374	2532	4834	84.28%	251.82%

查询操作比例	正常查询吞吐量/ （指令数·s^-1）	优化查询1吞吐量/ （指令数·s^-1）	优化查询2吞吐量/ （指令数·s^-1）	性能提升1	性能提升2
100%	1032	3931	8194	280.91%	693.99%
90%	1210	3553	7102	193.64%	486.94%
50%	1270	2540	4869	100.00%	283.39%

查询操作比例	正常查询吞吐量/ （指令数·s^-1）	优化查询1吞吐量/ （指令数·s^-1）	优化查询2吞吐量/ （指令数·s^-1）	性能提升1	性能提升2
100%	1014	3942	8153	288.76%	704.04%
90%	1205	3490	7204	189.63%	497.84%
50%	1374	2510	4851	82.68%	253.06%