2019
DOI: 10.1007/s00778-019-00581-w
|View full text |Cite
|
Sign up to set email alerts
|

In-memory database acceleration on FPGAs: a survey

Abstract: While FPGAs have seen prior use in database systems, in recent years interest in using FPGA to accelerate databases has declined in both industry and academia for the following three reasons. First, specifically for in-memory databases, FPGAs integrated with conventional I/O provide insufficient bandwidth, limiting performance. Second, GPUs, which can also provide high throughput, and are easier to program, have emerged as a strong accelerator alternative. Third, programming FPGAs required developers to have f… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
26
0
1

Year Published

2020
2020
2023
2023

Publication Types

Select...
3
3
1

Relationship

1
6

Authors

Journals

citations
Cited by 69 publications
(27 citation statements)
references
References 104 publications
0
26
0
1
Order By: Relevance
“…Each dataset consists of several benchmarks with cardinalities ranging from 2 10 to 2 22 unique keys. The relation size in all of the experiments was 256 million tuples (in line with previous research [59]).…”
Section: Dataset Descriptionmentioning
confidence: 99%
See 1 more Smart Citation
“…Each dataset consists of several benchmarks with cardinalities ranging from 2 10 to 2 22 unique keys. The relation size in all of the experiments was 256 million tuples (in line with previous research [59]).…”
Section: Dataset Descriptionmentioning
confidence: 99%
“…(2) Irregular data-flow where indirection in the memory access patterns breaks the data locality and hence causes cache misses. Some database operators, such as selection, exhibit control flow irregularity, while others, like hash-join and (hash-based) group-by aggregation, can demonstrate both [22].…”
Section: Introductionmentioning
confidence: 99%
“…Veritabanı sistemleri genellikle yüksek işlem gücü gerektiren komutlardan öte, çok sayıda komutun bir sıraya alınarak en kısa zamanda işlenerek yanıt verilmesi odağında çalışmak- (Fang et al, 2020). Her ne kadar veritabanı ilkelerine karşı durulan noktalar olsa da OLAP sayesinde işletmeler veri ambarlarından çok daha hızlı ve faydalı raporlar elde edebilmektedirler.…”
Section: Donanım Destekli Veritabanlarıunclassified
“…Many recent studies consider improving the speed of lossless decompression. The study in [12] discusses some of the prior work in the context of databases. To address block boundary problems, [19] explores the blocklevel parallelism by performing pattern matching on the delimiters to predict the block boundaries.…”
Section: Related Workmentioning
confidence: 99%