A distributed data exchange engine for polystores
-
Abdulrahman Kaitoua
Abdulrahman Kaitoua is a Senior big data architect and a team lead in the innovation and research team of GK-Software SE company in Berlin, Germany. He received his Ph. D. with honor in Information Technology from Politecnico di Milano in 2017.
,
Tilmann Rabl
Tilmann Rabl is a full professor and Chair of the Data Engineering Systems Group at Hasso Plattner Institute and the University of Potsdam. He is also cofounder of the startup bankmark.
Volker Markl is a Full Professor and Chair of the DIMA Group at TU Berlin and an Adjunct Full Professor at the University of Toronto. He is Director of the Intelligent Analytics for Massive Data Research Group at DFKI and Director of the Berlin Big Data Center.
Abstract
There is an increasing interest in fusing data from heterogeneous sources. Combining data sources increases the utility of existing datasets, generating new information and creating services of higher quality. A central issue in working with heterogeneous sources is data migration: In order to share and process data in different engines, resource intensive and complex movements and transformations between computing engines, services, and stores are necessary.
Muses is a distributed, high-performance data migration engine that is able to interconnect distributed data stores by forwarding, transforming, repartitioning, or broadcasting data among distributed engines’ instances in a resource-, cost-, and performance-adaptive manner. As such, it performs seamless information sharing across all participating resources in a standard, modular manner. We show an overall improvement of 30 % for pipelining jobs across multiple engines, even when we count the overhead of Muses in the execution time. This performance gain implies that Muses can be used to optimise large pipelines that leverage multiple engines.
Funding statement: This work has been supported through grants by the German Ministry for Education and Research as s BIFOLD (01IS18025A and 01IS18037).
About the authors
Abdulrahman Kaitoua is a Senior big data architect and a team lead in the innovation and research team of GK-Software SE company in Berlin, Germany. He received his Ph. D. with honor in Information Technology from Politecnico di Milano in 2017.
Tilmann Rabl is a full professor and Chair of the Data Engineering Systems Group at Hasso Plattner Institute and the University of Potsdam. He is also cofounder of the startup bankmark.
Volker Markl is a Full Professor and Chair of the DIMA Group at TU Berlin and an Adjunct Full Professor at the University of Toronto. He is Director of the Intelligent Analytics for Massive Data Research Group at DFKI and Director of the Berlin Big Data Center.
References
1. Hadoop Yarn, https://hadoop.apache.org/docs/current/hadoop-yarn/hadoop-yarn-site/YARN.html.Suche in Google Scholar
2. SciDB array DataBase, http://www.paradigm4.com/.Suche in Google Scholar
3. DistCp tool, Apache Hadoop, https://hadoop.apache.org/docs/current3/hadoop-distcp/DistCp.html.Suche in Google Scholar
4. Haynes, Brandon and Cheung, Alvin and Balazinska, Magdalena, PipeGen: Data Pipe Generator for Hybrid Analytics, Proceedings of the Seventh ACM Symposium on Cloud Computing (SOCC), 2016 Oct 5 (pp. 470–483). ACM.10.1145/2987550.2987567Suche in Google Scholar
5. Dziedzic, Adam and Elmore, Aaron J and Stonebraker, Michael Data transformation and migration in polystores, IEEE High Performance Extreme Computing Conference (HPEC) 2016 Sep 13 (pp. 1–6). IEEE.10.1109/HPEC.2016.7761594Suche in Google Scholar
6. J. Duggan, A. Elmore J, M. Stonebraker, M. Balazinska, B. Howe, J. Kepner, S. Madden, D. Maier, T. Mattson, S. Zdonik, The bigdawg polystore system, ACM Sigmod Record. 2015 Aug 12; 44(2):11–16.10.1145/2814710.2814713Suche in Google Scholar
7. Apache Arrow, https://arrow.apache.org/.Suche in Google Scholar
8. AKKA, http://akka.io/.Suche in Google Scholar
9. Apache Flink, https://flink.apache.org/.Suche in Google Scholar
10. Apache Spark, https://spark.apache.org/.Suche in Google Scholar
11. Encode: Encyclopedia of DNA Elements, https://www.encodeproject.org/.Suche in Google Scholar
12. H. Lim, Y. Han, S. Babu, How to Fit when No One Size Fits., CIDR 2013; 4:35.Suche in Google Scholar
13. K. Shvachko, H. Kuang, S. Radia, R. Chansler, The hadoop distributed file system, MSST 2010 May 3; 10:1–10.10.1109/MSST.2010.5496972Suche in Google Scholar
14. A. Kaitoua, P. Pinoli, M. Bertoni, S. Ceri, Framework for supporting genomic operations, IEEE Transactions on Computers. 2016 Aug 29; 66(3):443–457.10.1109/TC.2016.2603980Suche in Google Scholar
15. Lo, Ming-Ling and Ravishankar, Chinya V, (Spatial hash-joins, ACM SIGMOD Record 1996 Jun 1; 25(2):247–258). ACM.10.1145/233269.233337Suche in Google Scholar
16. M. Bertoni, S. Ceri, A. Kaitoua, P. Pinoli, Evaluating cloud frameworks on genomic applications, IEEE International Conference on Big Data (Big Data) 2015 Oct 29 (pp. 193–202). IEEE.10.1109/BigData.2015.7363756Suche in Google Scholar
17. A. Kaitoua, T. Rabl, A. Katsifodimos, V. Markl, Muses: Distributed Data Migration System for Polystores, IEEE 35th International Conference on Data Engineering (ICDE) 2019 Apr 8 (pp. 1602–1605).10.1109/ICDE.2019.00152Suche in Google Scholar
18. D. Agrawal, S. Chawla, A. Elmagarmid, Z. Kaoudi, M. Ouzzani, P. Papotti, J. Quiané-Ruiz, N. Tang, M. Zaki, Road to Freedom in Big Data Analytics, EDBT 2016 Jan 1 (pp. 479–484).Suche in Google Scholar
19. A. Jindal, J. Quiané-Ruiz, S. Madden, CARTILAGE: adding flexibility to the Hadoop skeleton, Proceedings of the 2013 ACM SIGMOD International Conference on Management of Data 2013 Jun 22 (pp. 1057–1060). ACM.10.1145/2463676.2465258Suche in Google Scholar
20. A. Gupta, V. Gadepally, M. Stonebraker, Cross-engine query execution in federated database systems, IEEE High Performance Extreme Computing Conference (HPEC) 2016 Sep 13 (pp. 1–6). IEEE.10.1109/HPEC.2016.7761648Suche in Google Scholar
21. S. Cattani, S. Ceri, A. Kaitoua, P. Pinoli, Evaluating Genomic Big Data Operations on SciDB and Spark, International Conference on Web Engineering 2017 Jun 5 (pp. 482–493). Springer, Cham.10.1007/978-3-319-60131-1_34Suche in Google Scholar
© 2020 Walter de Gruyter GmbH, Berlin/Boston
Artikel in diesem Heft
- Frontmatter
- Editorial
- Algorithms for Big Data
- Regular Papers
- Dictionary learning for transcriptomics data reveals type-specific gene modules in a multi-class setting
- Large-scale graph generation: Recent results of the SPP 1736 – Part II
- A distributed data exchange engine for polystores
- Feature-aware forecasting of large-scale time series data sets
- Optimization frameworks for machine learning: Examples and case study
- Solving subset sum with small space – Handling cryptanalytic Big Data
- Scaling up network centrality computations – A brief overview
Artikel in diesem Heft
- Frontmatter
- Editorial
- Algorithms for Big Data
- Regular Papers
- Dictionary learning for transcriptomics data reveals type-specific gene modules in a multi-class setting
- Large-scale graph generation: Recent results of the SPP 1736 – Part II
- A distributed data exchange engine for polystores
- Feature-aware forecasting of large-scale time series data sets
- Optimization frameworks for machine learning: Examples and case study
- Solving subset sum with small space – Handling cryptanalytic Big Data
- Scaling up network centrality computations – A brief overview
