Scalable Filtering Modules for Database Acceleration on FPGAs

Manev, Kristiyan; Vaishnav, Anuj; Kritikakis, Charalampos; Koch, Dirk

doi:10.1145/3337801.3337810

Cited by 8 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To alleviate the overhead of full reconfiguration, the relational database community pursued partial reconfiguration (Fig. 12(b)) where only parts of the accelerator architecture are switched (e. g., [41,85,95,131,153]). While this works for coarse-grained functionality switching during runtime, the part that is reconfigured still is unavailable for seconds.…”

Section: Accelerator Placementmentioning

confidence: 99%

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

Dann¹,

Ritter²,

Fröning³

2020

Preprint

View full text Add to dashboard Cite

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 analysis, and medical applications, due to their data model variety and scalability. The broad data variety and sheer size of datasets pose unique challenges for the system design and runtime (incl. power consumption). While CPU performance scaling becomes increasingly more difficult, we argue that NRDS can benefit from adding field programmable gate arrays (FPGAs) as accelerators. However, FPGA-accelerated NRDS have not been systematically studied, yet.To facilitate understanding of this emerging domain, we explore the fit of FPGA acceleration for NRDS with a focus on data model variety. We define the term NRDS class as a group of non-relational database systems supporting the same data model. This survey describes and categorizes the inherent differences and nontrivial trade-offs of relevant NRDS classes as well as their commonalities in the context of common design decisions when building such a system with FPGAs. For example, we found in the literature that for keyvalue stores the FPGA should be placed into the system as a smart network interface card (SmartNIC) to benefit from direct access of the FPGA to the network. However, more complex data models and processing of other classes (e. g., graph and document) commonly require more elaborate near-data or socket accelerator placements where the FPGA respectively has the only

show abstract

Section: Accelerator Placementmentioning

confidence: 99%

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

Dann¹,

Ritter²,

Fröning³

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…To enable resource elastic capabilities to full extend, we define a custom interface between modules that allows modules to communicate by streaming data between them on virtual channels. Our custom interface is an updated version of the one described in [5]. Additions to the interface specification include: 1) fields for sharing data intermediate values between modules, 2) protocol for modules to have direct access to external memory, 3) protocol for modules to have memory mapped registers in a global address space.…”

Section: Resource Elastic Stream Processing a Overviewmentioning

confidence: 99%

“…Filter modules (where clauses) that use Disjunctive Normal Form (DNF) representation for Boolean evaluation targeting Xilinx UltraScale+ are presented in [5]. By providing three generic filter modules (DNF8, DNF16, DNF32), all filter operations can be implemented with little resource overhead for arbitrary resource requirements.…”

Section: B Resource Elasticitymentioning

confidence: 99%

See 1 more Smart Citation

Resource Elastic Database Acceleration

Manev

Koch

2020

2020 30th International Conference on Field-Programmable Logic and Applications (FPL)

Self Cite

View full text Add to dashboard Cite

Database sizes are growing faster than the processing power in the post-Moore era due to the advent of big data applications, which make hardware acceleration mandatory. We propose dynamic stream processing using partial reconfiguration to provide a high-performance solution to runtime-known problems. This work researches the benefits of applying resource elastic techniques when building the execution pipeline.

show abstract