Performance analysis and testing of HBase based on its architecture

Cai, Lizhi; Huang, Shidong; Chen, Leilei; Zheng, Yang

doi:10.1109/icis.2013.6607866

Cited by 12 publications

(7 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite the fact that CNSSB is a user understandable organization, any OLAP query that involves at least one dimension of the DW will need to process two CFs, one for the fact's attributes and one for the aggregated dimensions. Our second implementation refers to the SameCF design and consists in storing all the attributes in a single CF, based on the DLA-CF-SSB (Dehdouh et al, 2015) and on the work of Cai et al (2013). This configuration aims to improve the performance of OLAP queries that analyze a greater number of dimensions, such as Scenario 2 (Section 1.1).…”

Section: Dim Dimmentioning

confidence: 99%

“…• Physically organize all attributes in the same CF, as proposed by Cai et al (2013) and presented as DLA-SSB (Dehdouh et al, 2015). We call this schema as SameCF.…”

Section: Proposed Investigationmentioning

confidence: 99%

“…All data stored in HBase is organized in column-families, as described in Section 2.2. As proposed by Cai et al (2013), there are an important feature that impact the performance of HBase's read and write operations: the attributes stored in the same or among different column-families.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Evaluation of the Challenges of Multilingualism in Data Warehouse Development

Dediź

Stanier

2016

Proceedings of the 18th International Conference on Enterprise Information Systems

View full text Add to dashboard Cite

ICEIS 2016 received 257 paper submissions from 42 countries in all continents, which makes it one of the largest conferences in the World in the area of Information Systems, thus demonstrating the success and global dimension of this conference. From these, 42 papers were selected for publication and presentation at the Conference as full papers. These numbers, leading to a full-paper acceptance ratio of 16%, show the intention of preserving a high quality forum for this conference, a quality that we intend to maintain in the future, for the next editions of this conference.The high number and high quality of the received papers imposed difficult choices in the selection process. To evaluate each submission, a double blind paper review was performed by the Program Committee, whose members are highly qualified researchers in ICEIS topic areas.All presented papers will be available at the SCITEPRESS Digital Library and will be submitted for indexation by Thomson Reuters Conference Proceedings Citation Index (ISI), INSPEC, DBLP, EI (Elsevier Index) and Scopus.Additionally, a short list of presented papers will be selected to be expanded into a forthcoming book of ICEIS 2016 Selected Papers to be published by Springer in the LNBIP Series.The technical program of the conference included a panel and 4 invited talks delivered by internationally distinguished speakers, namely: Claudia Loebbecke (University of Cologne, Germany), Sergio Gusmeroli (TXT e-solutions SPA, Italy), Wil Van Der Aalst (Technische Universiteit Eindhoven, Netherlands) and Jan Vom Brocke (University of Liechtenstein, Liechtenstein). Their participation positively contributes to reinforce the overall quality of the Conference and to provide a deeper understanding of the fields addressed by the conference.Moreover, ICEIS 2016 had a Doctoral Consortium on Enterprise Information Systems and 1 tutorial. We are thankful to the Conference Co-chairs (Olivier Camp and José Cordeiro) and Program Co-chairs (Slimane Hammoudi, Leszek Maciaszek and Michele M. Missikoff) for their dedication and hard work in organizing these events.We sincerely thank all the authors for their submissions and participation in ICEIS 2016. Furthermore, we would like to thank all the members of the program committee and reviewers, who helped us with their expertise, dedication and time. We would also like to thank the invited speakers for their excellent contribution in sharing their knowledge and vision and the workshop/special session chairs whose collaboration with ICEIS 2015 was much appreciated. Finally, we gratefully acknowledge the professional support of the ICEIS 2016 team for all organizational processes. XIIIWe hope that all colleagues find this a fruitful and inspiring conference. We hope to contribute to the development of the Enterprise Information Systems community and look forward to having additional research results presented at the next edition of ICEIS, details of which are available at http://www.iceis.org. Slimane Hammoudi University of Cologne, GermanyAbstract:...

show abstract

Section: Dim Dimmentioning

confidence: 99%

“…• Physically organize all attributes in the same CF, as proposed by Cai et al (2013) and presented as DLA-SSB (Dehdouh et al, 2015). We call this schema as SameCF.…”

Section: Proposed Investigationmentioning

confidence: 99%

See 1 more Smart Citation

An Evaluation of the Challenges of Multilingualism in Data Warehouse Development

Dediź

Stanier

2016

Proceedings of the 18th International Conference on Enterprise Information Systems

View full text Add to dashboard Cite

show abstract

“…Related to the performance evaluation of columnar NoSQL, the work of Cai et al (2013) measures the HBase performance by using two physical DW designs: (i) only one column-family with multiple columns; and (ii) multiple column-families, where every column-family has only one column. In case (i), reading one row means reading the data of all columns, even if the user does not need them.…”

Section: Related Workmentioning

confidence: 99%

Physical Data Warehouse Design on NoSQL Databases - OLAP Query Processing over HBase

Scabora

Brito

Ciferri

et al. 2016

Proceedings of the 18th International Conference on Enterprise Information Systems

View full text Add to dashboard Cite

Abstract:Nowadays, data warehousing and online analytical processing (OLAP) are core technologies in business intelligence and therefore have drawn much interest by researchers in the last decade. However, these technologies have been mainly developed for relational database systems in centralized environments. In other words, these technologies have not been designed to be applied in scalable systems such as NoSQL databases. Adapting a data warehousing environment to NoSQL databases introduces several advantages, such as scalability and flexibility. This paper investigates three physical data warehouse designs to adapt the Star Schema Benchmark for its use in NoSQL databases. In particular, our main investigation refers to the OLAP query processing over column-oriented databases using the MapReduce framework. We analyze the impact of distributing attributes among column-families in HBase on the OLAP query performance. Our experiments showed how processing time of OLAP queries was impacted by a physical data warehouse design regarding the number of dimensions accessed and the data volume. We conclude that using distinct distributions of attributes among column-families can improve OLAP query performance in HBase and consequently make the benchmark more suitable for OLAP over NoSQL databases.

show abstract

“…In the literature, studies have investigated how the star schema should be mapped into wide-column and documents NoSQL databases (BOUSSAHOUA; BOUSSAID; BENTAYEB, 2017;CHEVALIER et al, 2015;CHEVALIER et al, 2016). Others have evaluated OLAP query performance on wide-column oriented databases (CAI et al, 2013;SCABORA et al, 2016). However, because of their design purposes, NoSQL databases can not provide high query performance.…”

Section: Use Of Nosql Databasesmentioning

confidence: 99%

Data Warehouses na era do Big Data: processamento eficiente de Junções Estrela no Hadoop

Brito¹

View full text Add to dashboard Cite

The era of Big Data is here: the combination of unprecedented amounts of data collected every day with the promotion of open source solutions for massively parallel processing has shifted the industry in the direction of data-driven solutions. From recommendation systems that help you find your next significant one to the dawn of self-driving cars, Cloud Computing has enabled companies of all sizes and areas to achieve their full potential with minimal overhead. In particular, the use of these technologies for Data Warehousing applications has decreased costs greatly and provided remarkable scalability, empowering business-oriented applications such as Online Analytical Processing (OLAP). One of the most essential primitives in Data Warehouses are the Star Joins, i.e. joins of a central table with satellite dimensions. As the volume of the database scales, Star Joins become unpractical and may seriously limit applications. In this thesis, we proposed specialized solutions to optimize the processing of Star Joins. To achieve this, we used the Hadoop software family on a cluster of 21 nodes. We showed that the primary bottleneck in the computation of Star Joins on Hadoop lies in the excessive disk spill and overhead due to network communication. To mitigate these negative effects, we proposed two solutions based on a combination of the Spark framework with either Bloom filters or the Broadcast technique. This reduced the computation time by at least 38%. Furthermore, we showed that the use of full scan may significantly hinder the performance of queries with low selectivity. Thus, we proposed a distributed Bitmap Join Index that can be processed as a secondary index with loose-binding and can be used with random access in the Hadoop Distributed File System (HDFS). We also implemented three versions (one in MapReduce and two in Spark) of our processing algorithm that uses the distributed index, which reduced the total computation time up to 88% for Star Joins with low selectivity from the Star Schema Benchmark (SSB). Because, ideally, the system should be able to perform both random access and full scan, our solution was designed to rely on a two-layer architecture that is framework-agnostic and enables the use of a query optimizer to select which approaches should be used as a function of the query. Due to the ubiquity of joins as primitive queries, our solutions are likely to fit a broad range of applications. Our contributions not only leverage the strengths of massively parallel frameworks but also exploit more efficient access methods to provide scalable and robust solutions to Star Joins with a significant drop in total computation time.

show abstract

Performance analysis and testing of HBase based on its architecture

Cited by 12 publications

References 3 publications

An Evaluation of the Challenges of Multilingualism in Data Warehouse Development

An Evaluation of the Challenges of Multilingualism in Data Warehouse Development

Physical Data Warehouse Design on NoSQL Databases - OLAP Query Processing over HBase

Data Warehouses na era do Big Data: processamento eficiente de Junções Estrela no Hadoop

Contact Info

Product

Resources

About