This article focuses on data storage and computing issues in cloud environments. With the advent of the big data era, cloud computing has become a key support for data storage and intelligent computing. This article analyzes the key challenges in cloud storage and computing, including data reliability, availability, and security, as well as the resource coordination scheduling problem between GPU and CPU in intelligent computing. In response to these challenges, this article proposes multiple innovative solutions: firstly, the DMcache extension architecture MapperX is designed, which improves the availability of storage systems and shortens disk recovery time through adaptive metadata bittrees; Secondly, a SwornDisk encrypted storage method combining LSM architecture and remote update encryption mechanism was proposed to enhance data security; Thirdly, the Elastic Scheduler framework has been developed to achieve elastic scheduling of GPU and CPU collaborative computing through local gradient accumulation algorithm; The fourth is the design of ParaX acceleration method, which optimizes the deep learning computing performance of multi-core CPUs through the "single instance single core" strategy. These methods have achieved significant performance improvements in experiments. Finally, this article looks forward to future research directions, including intelligent ABT highly available collaborative storage mechanisms, highly readable SwornDisk technology, and GPU-CPU collaborative elastic computing methods combining ES and ParaX.

Yiru Zhang. Efficient Storage and Access Management System for Network Databases Driven by Cloud Computing Technology. Academic Journal of Engineering and Technology Science (2024) Vol. 7, Issue 6: 123-127. https://doi.org/10.25236/AJETS.2024.070618.

[1] Wang S, Lu Z, Cao Q, et al. Exploration and Exploitation for Buffer-Controlled HDD-Writes for SSD-HDD Hybrid Storage Server [J]. Transactions on Storage (TOS), 2022.

[2] Spillane R P, Wang W, Gao J, et al. Using an LSM tree file structure for the on-disk format of an object storage platform:US201816213714[P]. US11093472B2 [2024-09-13].

[3] Karame G, Soriente C. Method And System For Performing Remote Attestation With A Gateway In The Context Of A Trusted Execution Environment (TEE):US201917435712 [P]. US2022156390A1 [2024-09-13].

[4] Yu Y, He W, Jin J, et al. iDNA-ABT: advanced deep learning model for detecting DNA methylation with adaptive features and transductive information maximization [J]. Bioinformatics, 2021. DOI:10.1093/bioinformatics/btab677.

[5] Chung M T, Weidendorfer J, Kranzlmueller F D. From reactive to proactive load balancing for task-based parallel applications in distributedmemory machines [J]. Concurrency and computation: practice and experience, 2023, 35(24): 7828-7829. DOI:10.1002/cpe.7828.

[6] Schwartz J, Kurniawati H, Hutter M. Combining a Meta-Policy and Monte-Carlo Planning for Scalable Type-Based Reasoning in Partially Observable Environments[J]. ArXiv, 2023, 6067. DOI:10.48550/arXiv.2306.06067.

[7] Albraikan A A, Maray M, Alotaibi, Faiz Abdullah Alnfiai, Mrim M. Kumar, ArunSayed, Ahmed. Bio-Inspired Artificial Intelligence with Natural Language Processing Based on Deceptive Content Detection in Social Networking [J]. Biomimetics, 2023, 8(6).

[8] Phan L X, Chishti M, Miller R J, et al. Systems and methods for varying elastic modulusappliances:US202117383286[P]. US2021347103A1 [2024-09-13]. DOI: US6964564 B2.