View in EDS

HORIZONTAL VS. VERTICAL SCALING IN MODERN DATABASE SYSTEMS: A COMPARATIVE ANALYSIS OF PERFORMANCE AND COST TRADE-OFFS

Saved in:
Bibliographic Details
Title: HORIZONTAL VS. VERTICAL SCALING IN MODERN DATABASE SYSTEMS: A COMPARATIVE ANALYSIS OF PERFORMANCE AND COST TRADE-OFFS
Authors: Researcher
Publisher Information: Zenodo, 2024.
Publication Year: 2024
Subject Terms: Database Scaling Strategies, Vertical vs. Horizontal Scaling, Distributed Database Systems, Cloud-Native Databases, Serverless Database Architecture
Description: This article presents a comprehensive analysis of vertical and horizontal scaling strategies in modern database systems, examining their performance implications, cost-effectiveness, and real-world applications. Through a detailed literature review and case study analysis, we explore the technical aspects, advantages, and limitations of each approach. Vertical scaling, characterized by increasing the resources of a single server, is contrasted with horizontal scaling, which involves distributing workloads across multiple nodes. Our research incorporates recent developments in hybrid scaling techniques and cloud-native solutions, providing insights into their potential to overcome traditional scaling limitations. We employ a multi-faceted comparative framework to evaluate these strategies across various performance metrics, cost considerations, and operational complexities. The article also investigates emerging trends such as serverless and autonomous databases, discussing their potential impact on future scaling paradigms. By synthesizing findings from academic research and industry practices, this article offers valuable guidance for database engineers and system architects in selecting and implementing optimal scaling strategies. Our analysis reveals that while each approach has distinct advantages, the choice of scaling strategy is highly context-dependent, influenced by factors such as workload characteristics, growth projections, and organizational constraints. This work contributes to the ongoing discourse on database scalability, providing a foundation for informed decision-making in the rapidly evolving landscape of data management systems.
Document Type: Article
Language: English
DOI: 10.5281/zenodo.13851050
DOI: 10.5281/zenodo.13851049
Rights: CC BY
Accession Number: edsair.doi.dedup.....2a2e7dea412e420c751546c19cb652d2
Database: OpenAIRE
Description
Abstract:This article presents a comprehensive analysis of vertical and horizontal scaling strategies in modern database systems, examining their performance implications, cost-effectiveness, and real-world applications. Through a detailed literature review and case study analysis, we explore the technical aspects, advantages, and limitations of each approach. Vertical scaling, characterized by increasing the resources of a single server, is contrasted with horizontal scaling, which involves distributing workloads across multiple nodes. Our research incorporates recent developments in hybrid scaling techniques and cloud-native solutions, providing insights into their potential to overcome traditional scaling limitations. We employ a multi-faceted comparative framework to evaluate these strategies across various performance metrics, cost considerations, and operational complexities. The article also investigates emerging trends such as serverless and autonomous databases, discussing their potential impact on future scaling paradigms. By synthesizing findings from academic research and industry practices, this article offers valuable guidance for database engineers and system architects in selecting and implementing optimal scaling strategies. Our analysis reveals that while each approach has distinct advantages, the choice of scaling strategy is highly context-dependent, influenced by factors such as workload characteristics, growth projections, and organizational constraints. This work contributes to the ongoing discourse on database scalability, providing a foundation for informed decision-making in the rapidly evolving landscape of data management systems.
DOI:10.5281/zenodo.13851050