Beschleunigung relationaler Datenbanken mittels laufzeitadaptiver FPGA Cluster (FPGA-Cluster DB)

Funded by:

  • Zentrales Innovationsprogramm Mittelstand (ZIM) des Bundesministeriums für Wirtschaft und Eneregie (BMWi) – Fördermodul Kooperationsprojekte

Runtime:

  • 2013 - 2015

Project Coordinator:

Research Associate:

  • Dipl. Inf. Stefan Werner

Motivation

Relational databases with large data volumes causes performance issues and have a negative user experience.

Goals and Procedure

The goal of this project is hardware acceleration of database operations in relational databases with large data volumes. This addresses the core issue of so-called “big data” scenarios by enabling database access and thus data processing to be accelerated without requiring major interventions in the customers’ data and analysis tools.

Innovation and Prospects

The technical implementation and innovation should be achieved using a combined hardware/software system that outsources time-consuming database operations to a scalable cluster of programmable logic devices (FPGAs, Field Programmable Gate Arrays). The FPGA-Cluster DB database management system to be developed is intended to process several queries at the same time, recognize time-consuming operations and outsource them to a cluster of FPGAs. The FPGAs only ever implement the circuit structures that are required for the requests currently being processed. This is achieved using partial dynamic reconfiguration, which makes it possible to change parts of the configuration of an FPGA and thus its implemented circuit structures at runtime. The currently required data paths are assembled from a set of pre-synthesized database operators according to the modular principle and written into the FPGA cluster.

In this context, the IFIS is working on the subproject "Integration of hardware accelerators into database systems", which deals with the process and database-related aspects of the project. The widely used open source database system MySQL is to be expanded so that the possibilities of hardware accelerators are taken into account when creating the operator graph and the corresponding operators are outsourced to the FPGA cluster. The optimization potential should also be exploited when processing parallel requests.

Project Partners

  • Institute of Information Systems (IFIS), University of Lübeck
  • Technical University of Dresden

Dissertation

Stefan Werner:
Hybrid Architecture for Hardware-accelerated Query Processing in Semantic Web Databases based on Runtime Reconfigurable FPGAs
eingereicht Oktober 2016, angenommen Januar 2017, mündl. Prüfung am 06.02.2017 (Doktorvater: Groppe), Thesis

Publications

2020

Dennis Heinrich, Stefan Werner, Christopher Blochwitz, Thilo Pionteck, and Sven Groppe,
Hardware-aided update acceleration in a hybrid Semantic Web database system, The Journal of Supercomputing , vol. 76, no. 10, pp. 7961--7984, 2020.
Datei: s11227-018-2462-y
Bibtex: BibTeX
@article{heinrich20hardware,
  author    = {Dennis Heinrich and Stefan Werner and Christopher Blochwitz and Thilo Pionteck and Sven Groppe},
  title     = {Hardware-aided update acceleration in a hybrid Semantic Web database system},
  journal   = {The Journal of Supercomputing},
  volume    = {76},
  number    = {10},
  pages     = {7961--7984},
  year      = {2020},
  url       = {https://doi.org/10.1007/s11227-018-2462-y}
}

2017

Christopher Blochwitz, Julian Wolff, Jan Moritz Joseph, Stefan Werner, Dennis Heinrich, Sven Groppe, and Thilo Pionteck,
Hardware-Accelerated Radix-Tree Based String Sorting for Big Data Applications, in 30th International Conference on Architecture of Computing Systems ({ARCS}), Vienna, Austria , 2017. pp. 47--58.
DOI:10.1007/978-3-319-54999-6\_4
Datei: 978-3-319-54999-6\_4
Dennis Heinrich, Stefan Werner, Christopher Blochwitz, Thilo Pionteck, and Sven Groppe,
Search \& Update Optimization of a B+ Tree in a Hardware aided Semantic Web Database System, in Proceedings of the 7th International Conference on Emerging Databases (EDB), Busan, Korea , Springer, 2017.
Datei: 978-981-10-6520-0_18
Stefan Werner, Dennis Heinrich, Thilo Pionteck, and Sven Groppe,
Semi-static operator graphs for accelerated query execution on FPGAs, Microprocessors and Microsystems - Embedded Hardware Design , vol. 53, pp. 178--189, 2017.
DOI:10.1016/j.micpro.2017.07.010
Datei: j.micpro.2017.07.010
Stefan Werner,
{Hybrid Architecture for Hardware-accelerated Query Processing in Semantic Web Databases based on Runtime Reconfigurable FPGAs}, Universit{\"a}t zu L{\"u}beck, 2017.
Datei: ediss1886.pdf
Bibtex: BibTeX
@PhdThesis{WernerPhD2017,
  author = {Stefan Werner},
  title =  {{Hybrid Architecture for Hardware-accelerated Query Processing in Semantic Web Databases based on Runtime Reconfigurable FPGAs}},
  school = {Universit{\"a}t zu L{\"u}beck},
  year =   {2017}
}

2016

Stefan Werner, Dennis Heinrich, Marc Stelzner, Volker Linnemann, Thilo Pionteck, and Sven Groppe,
Accelerated join evaluation in Semantic Web databases by using FPGAs, Concurrency and Computation: Practice and Experience , vol. 28, no. 7, pp. 2031--2051, 2016.
DOI:10.1002/cpe.3502
Datei: cpe.3502
Bibtex: BibTeX
@ARTICLE{Werner2015,
  author = {Stefan Werner and Dennis Heinrich and Marc Stelzner and Volker Linnemann
	and Thilo Pionteck and Sven Groppe},
  title = {Accelerated join evaluation in Semantic Web databases by using FPGAs},
  journal = {Concurrency and Computation: Practice and Experience},
   year = {2015},
  volume =    {28},
  number =    {7},
  pages =     {2031--2051},
  month = {May 18},
  doi = {http://dx.doi.org/10.1002/cpe.3502}
}
Stefan Werner, Dennis Heinrich, Sven Groppe, Christopher Blochwitz, and Thilo Pionteck,
Runtime Adaptive Hybrid Query Engine based on FPGAs, OJDB , vol. 3, no. 1, pp. 21--41, 2016. RonPub.
Datei: urn:nbn:de:101:1-201705194645
Bibtex: BibTeX
@Article{OJDB_2016v3i1n02_Werner,
        author    = {Stefan Werner and Dennis Heinrich and Sven Groppe and Christopher Blochwitz and Thilo Pionteck},
        title     = {Runtime Adaptive Hybrid Query Engine based on FPGAs},
        journal   = {Open Journal of Databases (OJDB)},
        issn      = {2199-3459},
        year      = {2016},
        volume    = {3},
        number    = {1},
        pages     = {21--41},
        url       = {http://www.ronpub.com/publications/OJDB_2016v3i1n02_Werner.pdf},
        publisher = {RonPub},
        bibsource = {RonPub},
        abstract = {This paper presents the fully integrated hardware-accelerated query engine for large-scale datasets in the context of Semantic Web databases. As queries are typically unknown at design time, a static approach is not feasible and not flexible to cover a wide range of queries at system runtime. Therefore, we introduce a runtime reconfigurable accelerator based on a Field Programmable Gate Array (FPGA), which transparently incorporates with the freely available Semantic Web database LUPOSDATE. At system runtime, the proposed approach dynamically generates an optimized hardware accelerator in terms of an FPGA configuration for each individual query and transparently retrieves the query result to be displayed to the user. During hardware-accelerated execution the host supplies triple data to the FPGA and retrieves the results from the FPGA via PCIe interface. The benefits and limitations are evaluated on large-scale synthetic datasets with up to 260 million triples as well as the widely known Billion Triples Challenge.}
    }

2015

Christopher Blochwitz, Jan Moritz Joseph, Thilo Pionteck, Rico Backasch, Stefan Werner, and Dennis Heinrich,
{An optimized Radix-Tree for hardware-accelerated index generation for Semantic Web Databases}, in International Conference on ReConFigurable Computing and FPGAs (ReConFig) , Cancun, Mexico , 2015.
Bibtex: BibTeX
@INPROCEEDINGS{Blochwitz2015,
  author = {Christopher Blochwitz and Jan Moritz Joseph and Thilo Pionteck and
	Rico Backasch and  Stefan Werner and Dennis Heinrich and Sven Groppe},
  title = {{An optimized Radix-Tree for hardware-accelerated index generation
	for Semantic Web Databases}},
  booktitle = {International Conference on ReConFigurable Computing and FPGAs (ReConFig)},
  year = {2015},
  address = {Cancun, Mexico},
  month = {December 7 - 9},
  owner = {werner},
  timestamp = {2016.01.26}
}
Rico Backasch, Gerald Hempel, Christopher Blochwitz, Stefan Werner, Sven Groppe, and Thilo Pionteck,
An architectural template for composing application specific datapaths at runtime, in International Conference on ReConFigurable Computing and FPGAs (ReConFig), Riviera Maya, Mexico , 2015. pp. 1--6.
DOI:10.1109/ReConFig.2015.7393300
Datei: ReConFig.2015.7393300
Christopher Blochwitz, Jan Moritz Joseph, Rico Backasch, Thilo Pionteck, Stefan Werner, Dennis Heinrich, and Sven Groppe,
An optimized radix-tree for hardware-accelerated dictionary generation for semantic web databases, in International Conference on ReConFigurable Computing and FPGAs (ReConFig), Riviera Maya, Mexico , 2015. pp. 1--7.
DOI:10.1109/ReConFig.2015.7393291
Datei: ReConFig.2015.7393291
Stefan Werner, Dennis Heinrich, Jannik Piper, Sven Groppe, Rico Backasch, Christopher Blochwitz, and Thilo Pionteck,
Automated composition and execution of hardware-accelerated operator graphs, in 10th International Symposium on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC), Bremen, Germany , 2015. pp. 1--8.
DOI:10.1109/ReCoSoC.2015.7238078
Datei: ReCoSoC.2015.7238078
Sven Groppe, Dennis Heinrich, and Stefan Werner,
Distributed Join Approaches for W3C-Conform {SPARQL} Endpoints, OJSW , vol. 2, no. 1, pp. 30--52, 2015.
Datei: urn:nbn:de:101:1-201705194910
Dennis Heinrich, Stefan Werner, Marc Stelzner, Christopher Blochwitz, Thilo Pionteck, and Sven Groppe,
Hybrid FPGA approach for a B+ tree in a Semantic Web database system, in 10th International Symposium on Reconfigurable Communication-centric Systems-on-Chip (ReCoSoC), Bremen, Germany , 2015. pp. 1--8.
DOI:10.1109/ReCoSoC.2015.7238093
Datei: ReCoSoC.2015.7238093
Sven Groppe, Dennis Heinrich, Stefan Werner, Christopher Blochwitz, and Thilo Pionteck,
PatTrieSort - External String Sorting based on Patricia Tries, OJDB , vol. 2, no. 1, pp. 36--50, 2015.
Datei: urn:nbn:de:101:1-201705194627

2014

Stefan Werner, Dennis Heinrich, Marc Stelzner, Sven Groppe, Rico Backasch, and Thilo Pionteck,
Parallel and Pipelined Filter Operator for Hardware-Accelerated Operator Graphs in Semantic Web Databases, in 14th IEEE International Conference on Computer and Information Technology (CIT), Xi'an, China , 2014. pp. 539--546.
DOI:10.1109/CIT.2014.162
Datei: CIT.2014.162
Sven Groppe, Johannes Blume, Dennis Heinrich, and Stefan Werner,
A Self-Optimizing Cloud Computing System for Distributed Storage and Processing of Semantic Web Data, OJCC , vol. 1, no. 2, pp. 1--14, 2014.
Datei: urn:nbn:de:101:1-201705194478
Sven Groppe, Thomas Kiencke, Stefan Werner, Dennis Heinrich, Marc Stelzner, and Le Gruenwald,
P-LUPOSDATE: Using Precomputed Bloom Filters to Speed Up SPARQL Processing in the Cloud, OJSW , vol. 1, no. 2, pp. 25--55, 2014.
Datei: urn:nbn:de:101:1-201705194858
Rico Backasch, Gerald Hempel, Stefan Werner, Sven Groppe, and Thilo Pionteck,
Identifying homogenous reconfigurable regions in heterogeneous FPGAs for module relocation, in International Conference on ReConFigurable Computing and FPGAs (ReConFig), Cancun, Mexico , 2014. pp. 1--6.
DOI:10.1109/ReConFig.2014.7032533
Datei: ReConFig.2014.7032533

2013

Sven Groppe, Björn Schütt, and Stefan Werner,
Eliminating the XML overhead in embedded XML languages, in Proceedings of the 28th Annual ACM Symposium on Applied Computing ({SAC}), Coimbra, Portugal , 2013. pp. 542--547.
DOI:10.1145/2480362.2480466
Datei: 2480362.2480466
Stefan Werner, Sven Groppe, Volker Linnemann, and Thilo Pionteck,
Hardware-accelerated join processing in large Semantic Web databases with FPGAs, in International Conference on High Performance Computing \& Simulation (HPCS), Helsinki, Finland , 2013. pp. 131--138.
DOI:10.1109/HPCSim.2013.6641403
Datei: HPCSim.2013.6641403

2011

Stefan Werner, Christoph Reinke, and Volker Linnemann,
{Adaptive Service Migration in Wireless Sensor Networks}, in Proceedings of the 12th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT-2011) , Gwangju, Korea: CPS, 2011. pp. 334--341.
DOI:http://www.computer.org/portal/web/csdl/doi/10.1109/PDCAT.2011.15
ISBN:978-0-7695-4564-6
Bibtex: BibTeX
@Inproceedings{WeReGr11,
  author	=   {Stefan Werner and Christoph Reinke and Sven Groppe and Volker Linnemann},
  title		=   {{Adaptive Service Migration in Wireless Sensor Networks}},
  booktitle	=   {Proceedings of the 12th International Conference on Parallel and Distributed
                 Computing, Applications and Technologies (PDCAT-2011)},
  editor	=   {},
  year		=   {2011},
  month		=   {October 20 - 22},
  address	=   {Gwangju, Korea},
  pages		=   {334--341},
  publisher	=   {CPS},
  isbn		=   {978-0-7695-4564-6},
  doi		=   {http://www.computer.org/portal/web/csdl/doi/10.1109/PDCAT.2011.15},
  bibsource	=   {IFIS, Universit{\"a}t zu L{\"u}beck},
  copyright	=   {IEEE}
}
Christoph Reinke, Nils Hoeller, Stefan Werner, and Volker Linnemann,
{Analysis and Comparison of Concurrency Control Protocols for Wireless Sensor Networks}, in Proceedings of the 3rd International Workshop on Performance Control in Wireless Sensor Networks (PWSN 2011) in conjunction with the 7th IEEE International Conference on Distributed Computing in Sensor Systems (IEEE DCOSS '11) , Barcelona, Spain: IEEE, 2011. pp. 1-6.
DOI:http://dx.doi.org/10.1109/DCOSS.2011.5982219
ISBN:978-1-4577-0512-0
Bibtex: BibTeX
@Inproceedings{ReHoWe11a,
  author	=   {Christoph Reinke and Nils Hoeller and Stefan Werner and Sven Groppe and Volker Linnemann},
  title		=   {{Analysis and Comparison of Concurrency Control Protocols for Wireless Sensor Networks}},
  booktitle	=   {Proceedings of the 3rd International Workshop on Performance Control in Wireless Sensor Networks (PWSN 2011) in conjunction with the 7th IEEE International Conference on Distributed Computing in Sensor Systems (IEEE DCOSS '11)},
  year		=   {2011},
  month		=   {June 29},
  address	=   {Barcelona, Spain},
  pages		=   {1-6},
  publisher	=   {IEEE},
  isbn		=   {978-1-4577-0512-0},
  doi		=   {http://dx.doi.org/10.1109/DCOSS.2011.5982219},
  keywords	=   {concurrency control, transaction processing, wireless sensor networks},
  bibsource	=   {IFIS, Universit{\"a}t zu L{\"u}beck},
  copyright	=   {IEEE}
}
Christoph Reinke, Nils Hoeller, Stefan Werner, and Volker Linnemann,
{Consistent Service Migration in Wireless Sensor Networks}, in Proceedings of the 2011 International Conference on Wireless and Optical Communications (ICWOC 2011 - including ICIME 2011) , Zhengzhou, China: IEEE, 2011. pp. 278--285.
ISBN:978-1-61284-239-4
Bibtex: BibTeX
@Inproceedings{ReHoWe11,
  author	=   {Christoph Reinke and Nils Hoeller and Stefan Werner and Sven Groppe and Volker Linnemann},
  title		=   {{Consistent Service Migration in Wireless Sensor Networks}},
  booktitle	=   {Proceedings of the 2011 International Conference on Wireless and Optical Communications (ICWOC 2011 - including ICIME 2011)},
  year		=   {2011},
  month		=   {May 21 - 22},
  address	=   {Zhengzhou, China},
  pages		=   {278--285},
  publisher	=   {IEEE},
  isbn		=   {978-1-61284-239-4},
  doi		=   {},
  keywords	=   {service migration, transaction processing, wireless sensor networks},
  bibsource	=   {IFIS, Universit{\"a}t zu L{\"u}beck},
  copyright	=   {IEEE}
}
Stefan Werner, Christoph Reinke, Sven Groppe, and Volker Linnemann,
Adaptive Service Migration in Wireless Sensor Networks, in 12th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT), Gwangju, Korea , 2011. pp. 334--341.
DOI:10.1109/PDCAT.2011.15
Datei: PDCAT.2011.15
Christoph Reinke, Nils Hoeller, Stefan Werner, Sven Groppe, and Volker Linnemann,
Analysis and comparison of concurrency control protocols for wireless sensor networks, in 7th IEEE International Conference and Workshops on Distributed Computing in Sensor Systems (DCOSS), Barcelona, Spain , 2011. pp. 1--6.
DOI:10.1109/DCOSS.2011.5982219
Datei: DCOSS.2011.5982219