Performance Evaluation of a Distributed Backend System Based on Load Balancing and Replication Strategies
Article Sidebar
Main Article Content
Abstract
Distributed systems have become the foundation of modern cloud services, where applications are expected to provide high availability and low response latency under continuously increasing workloads. However, uneven traffic distribution and single-node failures can significantly reduce system performance and service reliability. This paper presents a distributed backend architecture that combines load balancing, data replication, and asynchronous task processing to improve system throughput and fault tolerance. The proposed system consists of four application nodes, two replicated database instances, and a Redis caching layer deployed in a containerized environment. Performance evaluations were conducted using Apache JMeter with concurrent workloads ranging from 100 to 2,000 virtual users. Experimental results show that the average response time increased from 82 ms under 100 concurrent users to 247 ms under 2,000 concurrent users, while the system maintained a 99.2% request success rate. Compared with a single-node deployment, the distributed architecture improved overall throughput by 41.8%, reduced average CPU utilization from 87% to 64%, and decreased database query latency by 35% through cache optimization. In addition, failover tests demonstrated that service recovery was completed within 5.8 seconds after a node failure, with no observable data loss. These results indicate that combining load balancing, replication, and caching can effectively improve the scalability, reliability, and operational efficiency of distributed backend systems.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Mind forge Academia also operates under the Creative Commons Licence CC-BY 4.0. This allows for copy and redistribute the material in any medium or format for any purpose, even commercially. The premise is that you must provide appropriate citation information.
References
Tanenbaum, A. S., & Van Steen, M. (2017). Distributed Systems: Principles and Paradigms (3rd ed.). Pearson.
Coulouris, G., Dollimore, J., Kindberg, T., & Blair, G. (2012). Distributed Systems: Concepts and Design (5th ed.). Addison-Wesley.
DeCandia, G., Hastorun, D., Jampani, M., Kakulapati, G., Lakshman, A., Pilchin, A., et al. (2007). Dynamo: Amazon's highly available key-value store. Proceedings of the 21st ACM SIGOPS Symposium on Operating Systems Principles (SOSP '07), 205–220. https://doi.org/10.1145/1294261.1294281
Corbett, J. C., Dean, J., Epstein, M., Fikes, A., Frost, C., Furman, J. J., et al. (2013). Spanner: Google's globally distributed database. ACM Transactions on Computer Systems, 31(3), Article 8. https://doi.org/10.1145/2491245
Ongaro, D., & Ousterhout, J. (2014). In Search of an Understandable Consensus Algorithm (Raft). Proceedings of the USENIX Annual Technical Conference (ATC), 305–319.
Gilbert, S., & Lynch, N. (2002). Brewer's conjecture and the feasibility of consistent, available, partition-tolerant web services. ACM SIGACT News, 33(2), 51–59. https://doi.org/10.1145/564585.564601
Vogels, W. (2009). Eventually Consistent. Communications of the ACM, 52(1), 40–44. https://doi.org/10.1145/1435417.1435432
Lakshman, A., & Malik, P. (2010). Cassandra: A decentralized structured storage system. ACM SIGOPS Operating Systems Review, 44(2), 35–40. https://doi.org/10.1145/1773912.1773922
Verma, A., Pedrosa, L., Korupolu, M., Oppenheimer, D., Tune, E., & Wilkes, J. (2015). Large-scale cluster management at Google with Borg. Proceedings of the Tenth European Conference on Computer Systems (EuroSys '15), Article 18. https://doi.org/10.1145/2741948.2741964
Barlocker, B., & Yan, X. (2021). Contrastive Representation Learning for Anomaly Detection in Distributed Backend Services. Artificial Intelligence and Computing Innovations, 1(2).