ETLMR: A Highly Scalable Dimensional ETL Framework based on MapReduce

Research output: Contribution to book/anthology/report/conference proceedingReport chapterResearch

19 Citations (Scopus)

Abstract

Extract-Transform-Load (ETL) flows periodically populate data warehouses (DWs) with data from different source systems. An increasing challenge for ETL fl ows is processing huge volumes of data quickly. MapReduce is establishing itself as the de-facto standard for large-scale data-intensive processing. However, MapReduce lacks support for high-level ETL specific constructs, resulting in low ETL programmer productivity. This report presents a scalable dimensional ETL framework, ETLMR, based on MapReduce. ETLMR has built-in native support for operations on DW-specific constructs such as star schemas, snowflake schemas and slowly changing dimensi ons (SCDs). This enables ETL developers to construct scalable MapReduce-based ETL flows with v ery few code lines. To achieve good performance and load balancing, a number of dimension and fact processing schemes are presented, including techniques for efficiently processing different types of dimensions. The report describes the integration of ETLMR with a MapReduce framework and evaluates its performance on large realistic data sets. The experimental results show that ETLMR achieves very good scalability and compares
favourably with other MapReduce data warehousing tools.
Original languageEnglish
Title of host publicationEnglish
Number of pages25
Place of PublicationTech Report TR-29
PublisherDepartment of Computer Science, Aalborg University
Publication date1 Aug 2011
Publication statusPublished - 1 Aug 2011

Fingerprint

Mathematical transformations
Data warehouses
Processing
Resource allocation
Stars
Scalability
Productivity

Bibliographical note

Technical Report

Cite this

Xiufeng, L., Thomsen, C., & Pedersen, T. B. (2011). ETLMR: A Highly Scalable Dimensional ETL Framework based on MapReduce. In English Tech Report TR-29: Department of Computer Science, Aalborg University.
Xiufeng, Liu ; Thomsen, Christian ; Pedersen, Torben Bach. / ETLMR: A Highly Scalable Dimensional ETL Framework based on MapReduce. English. Tech Report TR-29 : Department of Computer Science, Aalborg University, 2011.
@inbook{2dea9bf5e3484c34a6b458d8ee28faf4,
title = "ETLMR: A Highly Scalable Dimensional ETL Framework based on MapReduce",
abstract = "Extract-Transform-Load (ETL) flows periodically populate data warehouses (DWs) with data from different source systems. An increasing challenge for ETL fl ows is processing huge volumes of data quickly. MapReduce is establishing itself as the de-facto standard for large-scale data-intensive processing. However, MapReduce lacks support for high-level ETL specific constructs, resulting in low ETL programmer productivity. This report presents a scalable dimensional ETL framework, ETLMR, based on MapReduce. ETLMR has built-in native support for operations on DW-specific constructs such as star schemas, snowflake schemas and slowly changing dimensi ons (SCDs). This enables ETL developers to construct scalable MapReduce-based ETL flows with v ery few code lines. To achieve good performance and load balancing, a number of dimension and fact processing schemes are presented, including techniques for efficiently processing different types of dimensions. The report describes the integration of ETLMR with a MapReduce framework and evaluates its performance on large realistic data sets. The experimental results show that ETLMR achieves very good scalability and comparesfavourably with other MapReduce data warehousing tools.",
author = "Liu Xiufeng and Christian Thomsen and Pedersen, {Torben Bach}",
note = "Technical Report",
year = "2011",
month = "8",
day = "1",
language = "English",
booktitle = "English",
publisher = "Department of Computer Science, Aalborg University",

}

Xiufeng, L, Thomsen, C & Pedersen, TB 2011, ETLMR: A Highly Scalable Dimensional ETL Framework based on MapReduce. in English. Department of Computer Science, Aalborg University, Tech Report TR-29.

ETLMR: A Highly Scalable Dimensional ETL Framework based on MapReduce. / Xiufeng, Liu; Thomsen, Christian; Pedersen, Torben Bach.

English. Tech Report TR-29 : Department of Computer Science, Aalborg University, 2011.

Research output: Contribution to book/anthology/report/conference proceedingReport chapterResearch

TY - GEN

T1 - ETLMR: A Highly Scalable Dimensional ETL Framework based on MapReduce

AU - Xiufeng, Liu

AU - Thomsen, Christian

AU - Pedersen, Torben Bach

N1 - Technical Report

PY - 2011/8/1

Y1 - 2011/8/1

N2 - Extract-Transform-Load (ETL) flows periodically populate data warehouses (DWs) with data from different source systems. An increasing challenge for ETL fl ows is processing huge volumes of data quickly. MapReduce is establishing itself as the de-facto standard for large-scale data-intensive processing. However, MapReduce lacks support for high-level ETL specific constructs, resulting in low ETL programmer productivity. This report presents a scalable dimensional ETL framework, ETLMR, based on MapReduce. ETLMR has built-in native support for operations on DW-specific constructs such as star schemas, snowflake schemas and slowly changing dimensi ons (SCDs). This enables ETL developers to construct scalable MapReduce-based ETL flows with v ery few code lines. To achieve good performance and load balancing, a number of dimension and fact processing schemes are presented, including techniques for efficiently processing different types of dimensions. The report describes the integration of ETLMR with a MapReduce framework and evaluates its performance on large realistic data sets. The experimental results show that ETLMR achieves very good scalability and comparesfavourably with other MapReduce data warehousing tools.

AB - Extract-Transform-Load (ETL) flows periodically populate data warehouses (DWs) with data from different source systems. An increasing challenge for ETL fl ows is processing huge volumes of data quickly. MapReduce is establishing itself as the de-facto standard for large-scale data-intensive processing. However, MapReduce lacks support for high-level ETL specific constructs, resulting in low ETL programmer productivity. This report presents a scalable dimensional ETL framework, ETLMR, based on MapReduce. ETLMR has built-in native support for operations on DW-specific constructs such as star schemas, snowflake schemas and slowly changing dimensi ons (SCDs). This enables ETL developers to construct scalable MapReduce-based ETL flows with v ery few code lines. To achieve good performance and load balancing, a number of dimension and fact processing schemes are presented, including techniques for efficiently processing different types of dimensions. The report describes the integration of ETLMR with a MapReduce framework and evaluates its performance on large realistic data sets. The experimental results show that ETLMR achieves very good scalability and comparesfavourably with other MapReduce data warehousing tools.

M3 - Report chapter

BT - English

PB - Department of Computer Science, Aalborg University

CY - Tech Report TR-29

ER -

Xiufeng L, Thomsen C, Pedersen TB. ETLMR: A Highly Scalable Dimensional ETL Framework based on MapReduce. In English. Tech Report TR-29: Department of Computer Science, Aalborg University. 2011