Home / Current Issue / Paper 1710320
Modern Data Warehousing Architectures for Real-Time Business Decision Making
Subject area: Science,Engineering and Technology · Area of research: Data Warehousing
Abstract
In an age where information is growing at a faster rate and the need to make business decisions in real-time is an absolute must, enterprises are increasingly turning to sophisticated data warehousing solutions as a means of facilitating real-time business decisions. This paper examines how data warehousing as an idea has evolved, first using the batch-oriented architecture, to hybrids and cloud-native architectures that support real-time analytics. Based on a systematic review of scholarly publications, whitepapers provided by industry, and the results of industry benchmarking, the paper will evaluate the architecture in terms of latency, scalability, cost-efficiency, and the need to make decisions. This has shown that cloud-native microservice warehouses with streaming architecture, such as Apache Kafka, and ELT support, achieve superior latency-performance costs compared to legacy systems in low-latency queries at scale. A case study of a mid-sized retail enterprise further shows how real-time analytics can potentially optimize inventory and improve customer responsiveness. We end with a summary of best-practice guidelines for selecting and deploying modern data warehouse architectures in various business scenarios, and emergent trends in serverless warehousing, datamesh, and ML-based data warehouse query optimization. The knowledge presented in this article is intended to help practitioners and researchers to implement the real-time BI in their business in order to generate lasting and high impacts.
Keywords
Data Warehousing, Real-Time Analytics, Cloud Computing, ETL, Business Intelligence
How to cite this paper
@article{1710320,
author = {Jyothi Swaroop Myneni},
title = {Modern Data Warehousing Architectures for Real-Time Business Decision Making},
journal = {Iconic Research And Engineering Journals},
year = {2024},
volume = {8},
number = {3},
pages = {970-978},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1710320.pdf},
abstract = {In an age where information is growing at a faster rate and the need to make business decisions in real-time is an absolute must, enterprises are increasingly turning to sophisticated data warehousing solutions as a means of facilitating real-time business decisions. This paper examines how data warehousing as an idea has evolved, first using the batch-oriented architecture, to hybrids and cloud-native architectures that support real-time analytics. Based on a systematic review of scholarly publications, whitepapers provided by industry, and the results of industry benchmarking, the paper will evaluate the architecture in terms of latency, scalability, cost-efficiency, and the need to make decisions. This has shown that cloud-native microservice warehouses with streaming architecture, such as Apache Kafka, and ELT support, achieve superior latency-performance costs compared to legacy systems in low-latency queries at scale. A case study of a mid-sized retail enterprise further shows how real-time analytics can potentially optimize inventory and improve customer responsiveness. We end with a summary of best-practice guidelines for selecting and deploying modern data warehouse architectures in various business scenarios, and emergent trends in serverless warehousing, datamesh, and ML-based data warehouse query optimization. The knowledge presented in this article is intended to help practitioners and researchers to implement the real-time BI in their business in order to generate lasting and high impacts.},
keywords = {Data Warehousing, Real-Time Analytics, Cloud Computing, ETL, Business Intelligence},
month = {September},
}