Filters








1,652 Hits in 4.3 sec

doppioDB 1.0: Machine Learning inside a Relational Engine

Gustavo Alonso, Zsolt István, Kaan Kara, Muhsen Owaida, David Sidler
2019 IEEE Data Engineering Bulletin  
While these results are still a preliminary, exploratory step, they illustrate the challenges to be tackled and the advantages of using hardware accelerators as a way to extend database functionality in  ...  In this paper we present doppioDB 1.0, an FPGA-enabled database engine incorporating FPGA-based machine learning operators into a main memory, columnar DBMS (MonetDB).  ...  Acknowledgements We like to thank Intel for the generous donation of the Xeon+FPGA v2 prototype.  ... 
dblp:journals/debu/AlonsoIKOS19 fatcat:6vxbetufyrddxpy7tidsltuov4

An in-kernel NOSQL cache for range queries using FPGA NIC

Korechika Tamura, Hiroki Matsutani
2016 2016 International Conference on FPGA Reconfiguration for General-Purpose Computing (FPGA4GPC)  
To make use of big data, various NOSQL data stores have been deployed, such as key-value stores and column-oriented stores.  ...  In this paper, we also propose two cache strategies of IKNC for column-oriented stores.  ...  A query compiler for FPGA-based database accelerator, called Glacier, is proposed in [7] .  ... 
doi:10.1109/fpga4gpc.2016.7518528 fatcat:nbagv3z6ejcdnoygy6eww37lxq

AxleDB: A novel programmable query processing platform on FPGA

Behzad Salami, Gorker Alp Malazgirt, Oriol Arcas-Abella, Arda Yurdakul, Nehir Sonmez
2017 Microprocessors and microsystems  
In this work, we present the design and implementation of AxleDB: An FPGA-based platform that enables fast query processing for database systems by melding novel database-specific accelerators with commercial-off-the-shelf  ...  (COTS) storage using modern interfaces, in a novel, unified, and a programmable environment.  ...  query accelerators, and simultaneously optimizes the SSD I/O performance, thanks to the data movement techniques that were used, such as direct attached, column-oriented data storage and database indexing  ... 
doi:10.1016/j.micpro.2017.04.018 fatcat:iwsyzrtoifhivn552ywwqk3ude

FPGA Acceleration for Big Data Analytics: Challenges and Opportunities

Joost Hoozemans, Johan Peltenburg, Fabian Nonnemacher, Akos Hadnagy, Zaid Al-Ars, H. Peter Hofstee
2021 IEEE Circuits and Systems Magazine  
While these advantages of FPGA accelerators hold promise, difficulties associated with programming and integration limit their use.  ...  FPGAs implementing dataflow-oriented architectures with high levels of (pipeline) parallelism can provide high application throughput, often providing high energy efficiency.  ...  In addition, the format uses a column-oriented layout as opposed to the row-oriented layout used in traditional systems.  ... 
doi:10.1109/mcas.2021.3071608 fatcat:ywvymvx6bvc4dhytbgjcoqov2u

Ibex

Louis Woods, Zsolt István, Gustavo Alonso
2014 Proceedings of the VLDB Endowment  
Besides increasing performance, Ibex also reduces energy consumption, as it uses an FPGA rather than conventional CPUs to implement the off-load engine.  ...  GROUP BY aggregation has a higher impact on performance but is also a more challenging operator to implement on an FPGA.  ...  SQL Acceleration with FPGAs Initial approaches that used FPGAs for database tasks were mostly targeted at stream processing, e.g., Glacier [21] is a compiler that translates SQL queries into VHDL code  ... 
doi:10.14778/2732967.2732972 fatcat:f6f4rtrd3ffinecxzwq6k3gpm4

Column Scan Acceleration in Hybrid CPU-FPGA Systems

Nusrat Jahan Lisa, Annett Ungethüm, Dirk Habich, Wolfgang Lehner, Tuan D. A. Nguyen, Akash Kumar
2018 Very Large Data Bases Conference  
Nowadays, in-memory column store database systems are state-of-the-art for analytical workloads.  ...  Based on that, we focus on column scan acceleration for hybrid hardware systems incorporating a Field Programmable Gate Array (FPGA) and a CPU into a single system in this paper.  ...  For instance, the database architecture shifted from a disk-oriented to a main memory-oriented architecture to efficiently exploit the everincreasing capacities of main memory [1, 22, 27, 37] .  ... 
dblp:conf/vldb/LisaUHLN018 fatcat:mf2a3ca2effh5mz4abppxiyaau

Non-Relational Databases on FPGAs: Survey, Design Decisions, Challenges [article]

Jonas Dann, Daniel Ritter, Holger Fröning
2020 arXiv   pre-print
Across the different classes, FPGAs can be used as communication layer or for acceleration of operators and data access.  ...  Non-relational database systems (NRDS), such as graph, document, key-value, and wide-column, have gained much attention in various trending (business) application domains like smart logistics, social network  ...  Wide-column Wide-column databases also store data as pairs of key and value. However, different to key-value databases, the key has two predefined parts: a row name and a column name.  ... 
arXiv:2007.07595v1 fatcat:eganwap76ffpzbl6fztprecawq

Symbiote

Pranav S. Vaidya, Jaehwan John Lee, Francis Bowen, Yingzi Du, Chandima H. Nadungodage, Yuni Xia
2010 Proceedings of the 2010 international conference on Management of data - SIGMOD '10  
data stream processing using Field Programmable Gate Arrays (FPGAs).  ...  designed Symbiote -a distributed Reconfigurable Logic Assisted multimedia Data Stream Management System (RLADSMS) at Indiana University Purdue University, Indianapolis (IUPUI) that provides hardware accelerated  ...  For example, we have developed a parallel hardware version of the join operator similar to our hardware join operator for column-oriented databases [16] .  ... 
doi:10.1145/1807167.1807304 dblp:conf/sigmod/VaidyaLBDNX10 fatcat:p6pe2enib5e33e577frco74mly

Relational Memory: Native In-Memory Accesses on Rows and Columns [article]

Shahin Roozkhosh, Denis Hoornaert, Ju Hyoung Mun, Tarikul Islam Papon, Ahmed Sanaullah, Ulrich Drepper, Renato Mancuso, Manos Athanassoulis
2022 arXiv   pre-print
Analytical database systems are typically designed to use a column-first data layout to access only the desired fields.  ...  Our approach, termed Relational Memory, relies on an FPGA- based accelerator that sits between the CPU and main memory and transparently transforms base data to any group of columns with minimal overhead  ...  For this reason, FPGA technology has been widely used to implement specialized accelerators.  ... 
arXiv:2109.14349v2 fatcat:wh7g3ezf6jg2pd7zbd7727xmy4

High Bandwidth Memory on FPGAs: A Data Analytics Perspective [article]

Kaan Kara, Christoph Hagleitner, Dionysios Diamantopoulos, Dimitris Syrivelis, Gustavo Alonso
2020 arXiv   pre-print
HBM promises overcoming the bandwidth bottleneck, faced often by FPGA-based accelerators due to their throughput oriented design.  ...  We consider three workloads that are often performed in analytics oriented databases and implement them on FPGA showing in which cases they benefit from HBM: range selection, hash join, and stochastic  ...  MonetDB [24] is a column-oriented in-memory database optimized for online analytical processing (OLAP).  ... 
arXiv:2004.01635v1 fatcat:wrme53ej3bgarbeeadw7vvgvpe

FPGAs

Rene Mueller, Jens Teubner
2010 Proceedings of the 13th International Conference on Extending Database Technology - EDBT '10  
With a focus on database use, this tutorial introduces into FPGA technology, demonstrates its potential, but also pinpoints some challenges that need to be addressed before FPGA-accelerated database systems  ...  The goal of this tutorial is to develop an intuition of an FPGA development cycle, receive guidelines for a "good" FPGA design, but also learn the limitations that hardware-implemented database processing  ...  The programmed chip can then be used, e.g., as a hardware-accelerated implementation for specific compute or control tasks.  ... 
doi:10.1145/1739041.1739137 dblp:conf/edbt/MullerT10 fatcat:i5moqwftu5acri624w52jmooiy

Heterogeneous Cloud Framework for Big Data Genome Sequencing

Chao Wang, Xi Li, Peng Chen, Aili Wang, Xuehai Zhou, Hong Yu
2015 IEEE/ACM Transactions on Computational Biology & Bioinformatics  
In this paper, we propose a novel FPGA-based acceleration solution with MapReduce framework on multiple hardware accelerators.  ...  Also, as a practical study, we have built a hardware prototype on the real Xilinx FPGA chip.  ...  The biggest difference between Hbase and other SQL database is that Hbase is column-orient.  ... 
doi:10.1109/tcbb.2014.2351800 pmid:26357087 fatcat:52gw5e6o6jc5tefrr3xau3kdcm

The Glass Half Full: Using Programmable Hardware Accelerators in Analytics

Zsolt István
2019 IEEE Data Engineering Bulletin  
There have been numeorus proposals to accelerate databases using specialized hardware in past years but often the opinion of the community is pessimistic: the performance and energy efficiency benefits  ...  In this paper we highlight several challenges that have traditionally hindered the deployment of hardware acceleration in databases and explain how they have been alleviated or removed altogether by recent  ...  Today, column-oriented databases, such as MonetDB [27] , are widely deployed and typically outperform row-oriented ones by at least an order of magnitude and can take advantage of many-core CPUs efficiently  ... 
dblp:journals/debu/Istvan19 fatcat:z7a3z66tzzg5xd4nizbz47guli

4.-8. März 2019

Andreas Becher, Achim Herrmann, Stefan Wildermann, Jürgen Teich
2019 Datenbanksysteme für Business, Technologie und Web  
Reconfigurable hardware such as Field-programmable Gate Arrays (FPGAs) is widely used for data processing in databases.  ...  This work focuses on the architectural aspects of database (co-)processing on heterogeneous FPGA-based PSoC (programmable System-on-Chip) architectures including processors, specialized hardware components  ...  Fig. 2 : A tuple (a, b, c) being processed according to the query SELECT c,a,(a+c); using a roworiented storage (a) and a column-oriented storage (b).  ... 
doi:10.18420/btw2019-ws-04 dblp:conf/btw/BecherHWT19 fatcat:nb5aozrb5zfk5mt3vfm5pyt5ly

Tens of gigabytes per second JSON-to-Arrow conversion with FPGA accelerators

Johan Peltenburg, Akos Hadnagy, Matthijs Brobbel, Robert Morrow, Zaid Al-Ars
2021 2021 International Conference on Field-Programmable Technology (ICFPT)  
Several fully integrated FPGA-accelerated JSON parser implementations are presented using the Intel Arria 10 GX and Xilinx VU37P devices, and compared to the performance of their respective host systems  ...  Due to performance limitation of the CPU and GPU implementations, we furthermore present an FPGA accelerated implementation.  ...  Now consider the column-oriented approach, where the values in a column are placed contiguously in a buffer.  ... 
doi:10.1109/icfpt52863.2021.9609833 fatcat:hxak43dcqrbj7pxnbxvlwt6kyq
« Previous Showing results 1 — 15 out of 1,652 results