Intelligent caching in high-load systems as a response to the limitations of classical cache management methods
DOI:
https://doi.org/10.32347/2707-501x.2023.52(3).227-236Keywords:
adaptability, data, data flow management, data storage, database, information system, query processing optimization, caching technologyAbstract
The relevance of the work is substantiated by the importance of optimal data processing in highly loaded systems, which is achieved by using caching technology. The problems of effective management of data flows in modern information systems are researched. The basic functions and main types of caching in high-load information systems are considered. The role of caching in ensuring the scalability and stability of highly loaded systems in the conditions of growing requirements for their reliability, adaptability, stability and performance is defined. It is specified that in the realities of modern dynamic, highly loaded distributed systems, there are more and more situations in which these methods become insufficient or ineffective. At the same time, increasing the efficiency of data access, load balancing, reducing latency and ensuring system stability directly depends on a properly organised caching mechanism. The typical limitations of classical cache management methods are analysed. It is shown that methods such as Least Recently Used, Least Frequently Used, First In First Out and their modifications are more often demonstrating limitations in a dynamic environment where changes in the structure of requests, the context of objects and variability of user behaviour play an important role. As a response to the limitations of classical cache management methods for tasks where it is essential to take into account the specifics of queries and user behaviour of different groups, the context and relationships between objects, as well as the dynamics of queries based on their history, the expediency of using intelligent cache management strategies is substantiated. The prospect of development and integration of intelligent cache management components into the operation of high-load systems is shown, which will allow them to learn from real data, predict future requests and make effective decisions on saving or deleting objects from the cache. It has been decided to focus further research on the development of an intelligent caching method that can provide efficient access to relevant data in the information infrastructure of systems in the field of architecture and construction.
References
Cisco Annual Internet Report (2018–2022). URL: https://www.cisco.com/c/en/us/solutions/executive-perspectives/annual-internet-report/index.html
Hazelcast. What Is Caching? How Caching Works and Its Limitations. URL: https://hazelcast.com/foundations/caching/caching/
Prisma. Database caching: Overview, types, strategies and their benefits. URL: https://www.prisma.io/dataguide/managing-databases/introduction-database-caching
Megiddo N., Modha D. ARC: A Self-Tuning, Low Overhead Replacement Cache, USENIX Conference on File and Storage Technologies (FAST), 2003. URL: https://www.usenix.org/ legacy/event/fast03/tech/full_papers/megiddo/megiddo.pdf
Rotem-Gal-Oz A. Fallacies of Distributed Computing Explained, 2006. URL: https://arnon.me/wp-content/uploads/Files/fallacies.pdf
Swain D., Paikaray B., Swain D. AWRP: Adaptive Weight Ranking Policy for Improving Cache Performance, Journal of Computing, Vol. 3 (2), 2011. DOI: 10.48550/arXiv.1107.4851
Scully T., Jiang S., Zhang X. Learning-based Cache Replacement for High Performance and Dynamic Workloads. Proceedings of the 2020 IEEE International Conference on Big Data. 2020, p. 4042-4051.
М Liu E., Hashemi M., Swersky K., Ranganathan P., Ahn J. An Imitation Learning Approach for Cache Replacement, International Conference on Machine Learning (ICML), 2020. DOI: https://doi.org/10.48550/arXiv.2006.16239
Gartner. Magic Quadrant for Distributed Data Management Solutions. URL: https://www.gartner.com/en/documents/4000163
CacheFly. The Power of Machine Learning for Advanced CDN Caching Strategies. 2022, URL: https://www.cachefly.com/news/the-power-of-machine-learning-for-advanced-cdn-caching-strategies
Cisco. Global Networking Trends Report, 2022. URL: https://storage.eventcheckin.co.kr/cisco/2023/CXO_symposium/data/2022_Global_Networking_Trend_Report_eng.pdf
Weiner M., Xu Y., "Performance Metrics for Distributed Systems", ACM Computing Surveys, 2022, Vol. 54, No. 7, P. 1–29.
Downloads
Published
How to Cite
Issue
Section
License

This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).