Centaur: A Framework for Hybrid CPU-FPGA Databases

Owaida, Muhsen; Sidler, David; Kara, Kaan; Alonso, Gustavo

doi:10.1109/fccm.2017.37

Cited by 66 publications

(54 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…On the Intel Xeon+FPGA platform, we used the Intel's AAL framework for the FPGA applications development. Atop of AAL, we used Centaur [41], a framework for integrating FPGA accelerators with software applications. Centaur offers concurrent access to multiple accelerators sharing the same FPGA fabric.…”

Section: Fpga-software Integrationmentioning

confidence: 99%

Lowering the latency of data processing pipelines through FPGA based hardware acceleration

et al. 2019

Self Cite

View full text Add to dashboard Cite

Web search engines often involve a complex pipeline of processing stages including computing, scoring, and ranking potential answers plus returning the sorted results. The latency of such pipelines can be improved by minimizing data movement, making stages faster, and merging stages. The throughput is determined by the stage with the smallest capacity and it can be improved by allocating enough parallel resources to each stage. In this paper we explore the possibility of employing hardware acceleration (an FPGA) as a way to improve the overall performance when computing answers to search queries. With a real use case as a baseline and motivation, we focus on accelerating the scoring function implemented as a decision tree ensemble, a common approach to scoring and classification in search systems. Our solution uses a novel decision tree ensemble implementation on an FPGA to: 1) increase the number of entries that can be scored per unit of time, and 2) provide a compact implementation that can be combined with previous stages. The resulting system, tested in Amazon F1 instances, significantly improves the quality of the search results and improves performance by two orders of magnitude over the existing CPU based solution.

show abstract

Section: Fpga-software Integrationmentioning

confidence: 99%

Lowering the latency of data processing pipelines through FPGA based hardware acceleration

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Accelerating database operations is a popular research direction that connects modern acceleration platforms and enterprise in-database analytics as shown in Figure 1. These prior FPGA-based solu-tions aim to accelerate DBMS operations (some portion of the query) [3,4,23,24,46], such as join and hash. LINQits [46] accelerates database queries but does not focus on machine learning.…”

Section: Related Workmentioning

confidence: 99%

“…LINQits [46] accelerates database queries but does not focus on machine learning. Centaur [3] dynamically decides which particular operators in a MonetDB [47] query plan can be executed on FPGA and creates a pipeline between FPGA and CPU. Another work [24] uses FPGAs to provide a robust hashing mechanism to accelerate data partitioning in database engines.…”

Section: Related Workmentioning

confidence: 99%

In-RDBMS hardware acceleration of advanced analytics

et al. 2018

View full text Add to dashboard Cite

The data revolution is fueled by advances in machine learning, databases, and hardware design. Programmable accelerators are making their way into each of these areas independently. As such, there is a void of solutions that enables hardware acceleration at the intersection of these disjoint fields. This paper sets out to be the initial step towards a unifying solution for in-Database Acceleration of Advanced Analytics (DAnA). Deploying specialized hardware, such as FPGAs, for in-database analytics currently requires hand-designing the hardware and manually routing the data. Instead, DAnA automatically maps a high-level specification of advanced analytics queries to an FPGA accelerator. The accelerator implementation is generated for a User Defined Function (UDF), expressed as a part of an SQL query using a Pythonembedded Domain-Specific Language (DSL). To realize an efficient in-database integration, DAnA accelerators contain a novel hardware structure, Striders, that directly interface with the buffer pool of the database. Striders extract, cleanse, and process the training data tuples that are consumed by a multi-threaded FPGA engine that executes the analytics algorithm. We integrate DAnA with PostgreSQL to generate hardware accelerators for a range of real-world and synthetic datasets running diverse ML algorithms. Results show that DAnA-enhanced PostgreSQL provides, on average, 8.3× end-to-end speedup for real datasets, with a maximum of 28.2×. Moreover, DAnA-enhanced PostgreSQL is, on average, 4.0× faster than the multi-threaded Apache MADLib running on Greenplum. DAnA provides these benefits while hiding the complexity of hardware design from data scientists and allowing them to express the algorithm in ≈30-60 lines of Python.

show abstract

“…Instead, we introduce our custom system call termed vm_snapshot and integrate the concept of rewiring [19] directly into the Linux kernel. Such a co-design of underlying system components and the DBMS has been demonstrated successfully in recent publications concerning both operating system [9,13] and hardware [17,21] customizations, as it enables a whole new level of optimization opportunities. Using our call, we can essentially snapshot arbitrary virtual memory areas within a single process at any point in time.…”

Section: High-frequency Snapshottingmentioning

confidence: 99%

Accelerating Analytical Processing in MVCC using Fine-Granular High-Frequency Virtual Snapshotting

Sharma

Schuhknecht

Dittrich

2018

Proceedings of the 2018 International Conference on Management of Data

View full text Add to dashboard Cite

Efficient transaction management is a delicate task. As systems face transactions of inherently different types, ranging from point updates to long-running analytical queries, it is hard to satisfy their requirements with a single execution engine. Unfortunately, most systems rely on such a design that implements its parallelism using multi-version concurrency control. While MVCC parallelizes shortrunning OLTP transactions well, it struggles in the presence of mixed workloads containing long-running OLAP queries, as scans have to work their way through vast amounts of versioned data. To overcome this problem, we reintroduce the concept of hybrid processing and combine it with state-of-the-art MVCC: OLAP queries are outsourced to run on separate virtual snapshots while OLTP transactions run on the most recent version of the database. Inside both execution engines, we still apply MVCC.The most significant challenge of a hybrid approach is to generate the snapshots at a high frequency. Previous approaches heavily suffered from the high cost of snapshot creation. In our approach termed AnKer, we follow the current trend of co-designing underlying system components and the DBMS, to overcome the restrictions of the OS by introducing a custom system call vm_snapshot. It allows fine-granular snapshot creation that is orders of magnitudes faster than state-of-the-art approaches. Our experimental evaluation on an HTAP workload based on TPC-C transactions and OLAP queries show that our snapshotting mechanism is more than a factor of 100x faster than fork-based snapshotting and that the latency of OLAP queries is up to a factor of 4x lower than MVCC in a single execution engine. Besides, our approach enables a higher OLTP throughput than all state-of-the-art methods.

show abstract

Centaur: A Framework for Hybrid CPU-FPGA Databases

Cited by 66 publications

References 24 publications

Lowering the latency of data processing pipelines through FPGA based hardware acceleration

Lowering the latency of data processing pipelines through FPGA based hardware acceleration

In-RDBMS hardware acceleration of advanced analytics

Accelerating Analytical Processing in MVCC using Fine-Granular High-Frequency Virtual Snapshotting

Contact Info

Product

Resources

About