Accelerating business analytics applications

Salapura, Valentina; Karkhanis, Tejas S.; Nagpurkar, Priya; Moreira, José E.

doi:10.1109/hpca.2012.6169044

Cited by 26 publications

(14 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The available operations treat the data as vectors and focus largely on arithmetic and shuffling operations on the vector values. Many programmers have retrofitted and vectorized other types of programs, notably text parsing [5,32] and regular expression matching [45] and database kernels [55,18,31]. Our experiments in Section 2.2 using a state of the art SIMD range partitioning [46] indicate that vector-based traversal improves throughput somewhat but fails to fully saturate DRAM bandwidth.…”

Section: Related Workmentioning

confidence: 97%

See 1 more Smart Citation

Navigating big data with high-throughput, energy-efficient data partitioning

Barker

Kim

et al. 2013

SIGARCH Comput. Archit. News

View full text Add to dashboard Cite

The global pool of data is growing at 2.5 quintillion bytes per day, with 90% of it produced in the last two years alone [24]. There is no doubt the era of big data has arrived. This paper explores targeted deployment of hardware accelerators to improve the throughput and energy efficiency of largescale data processing. In particular, data partitioning is a critical operation for manipulating large data sets. It is often the limiting factor in database performance and represents a significant fraction of the overall runtime of large data queries.To accelerate partitioning, this paper describes a hardware accelerator for range partitioning, or HARP, and a hardware-software data streaming framework. The streaming framework offers a seamless execution environment for streaming accelerators such as HARP. Together, HARP and the streaming framework provide an order of magnitude improvement in partitioning performance and energy. A detailed analysis of a 32nm physical design shows 7.8 times the throughput of a highly optimized and optimistic software implementation, while consuming just 6.9% of the area and 4.3% of the power of a single Xeon core in the same technology generation.

show abstract

Section: Related Workmentioning

confidence: 97%

“…One recent study of regular expression matching compared different strategies for acceleration [45]. The study concluded that SIMD software was the best option, due to the fast data and control transfers between the scalar CPU and the vector unit.…”

Section: Related Workmentioning

confidence: 99%

Navigating big data with high-throughput, energy-efficient data partitioning

Barker

Kim

et al. 2013

SIGARCH Comput. Archit. News

View full text Add to dashboard Cite

show abstract

“…GPUs are perhaps the most visible and among the most successful such processors targeting graphics applications [5,30]. There is a large body of research around other domain-specific acceleration: Convolution Engine [32] targets image processing kernels and stencil computations, [33] uses specialization to speed up regular expression matching in queries, and [18] accelerates H.264 video encoders. In spirit these projects share similarities with the Q100, but in their design and target particulars they are quite different.…”

Section: Related Workmentioning

confidence: 99%

Q100

Lottarini

Paine

et al. 2014

Proceedings of the 19th International Conference on Architectural Support for Programming Languages and Operating Systems

120

View full text Add to dashboard Cite

In this paper, we propose Database Processing Units, or DPUs, a class of domain-specific database processors that can efficiently handle database applications. As a proof of concept, we present the instruction set architecture, microarchitecture, and hardware implementation of one DPU, called Q100. The Q100 has a collection of heterogeneous ASIC tiles that process relational tables and columns quickly and energy-efficiently. The architecture uses coarse grained instructions that manipulate streams of data, thereby maximizing pipeline and data parallelism, and minimizing the need to time multiplex the accelerator tiles and spill intermediate results to memory. This work explores a Q100 design space of 150 configurations, selecting three for further analysis: a small, power-conscious implementation, a highperformance implementation, and a balanced design that maximizes performance per Watt. We then demonstrate that the power-conscious Q100 handles the TPC-H queries with three orders of magnitude less energy than a state of the art software DBMS, while the performance-oriented design outperforms the same DBMS by 70X.

show abstract

“…The available operations treat the data as vectors and focus largely on arithmetic and shuffling operations on the vector values. Many programmers have retrofitted and vectorized other types of programs, notably text parsing [Cameron and Lin 2009;Lin et al 2012] and regular expression matching [Salapura et al 2012] and database kernels [Zhou and Ross 2002;Govindaraju and Manocha 2005;Krueger et al 2011]. Our experiments in Section 3 using a state-of-the-art SIMD range partitioning [Polychroniou and Ross 2014] indicate that vector-based traversal improves throughput and energy markedly.…”

Section: Related Workmentioning

confidence: 99%

“…One recent study of regular expression matching compared different strategies for acceleration [Salapura et al 2012]. The study concluded that SIMD software was the best option due to the fast data and control transfers between the scalar CPU and the vector unit.…”

Section: Related Workmentioning

confidence: 99%

Energy Analysis of Hardware and Software Range Partitioning

Polychroniou

Barker

et al. 2014

ACM Trans. Comput. Syst.

View full text Add to dashboard Cite

Data partitioning is a critical operation for manipulating large datasets because it subdivides tasks into pieces that are more amenable to efficient processing. It is often the limiting factor in database performance and represents a significant fraction of the overall runtime of large data queries. This article measures the performance and energy of state-of-the-art software partitioners, and describes and evaluates a hardware range partitioner that further improves efficiency.The software implementation is broken into two phases, allowing separate analysis of the partition function computation and data shuffling costs. Although range partitioning is commonly thought to be more expensive than simpler strategies such as hash partitioning, our measurements indicate that careful data movement and optimization of the partition function can allow it to approach the throughput and energy consumption of hash or radix partitioning.For further acceleration, we describe a hardware range partitioner, or HARP, a streaming framework that offers a seamless execution environment for this and other streaming accelerators, and a detailed analysis of a 32nm physical design that matches the throughput of four to eight software threads while consuming just 6.9% of the area and 4.3% of the power of a Xeon core in the same technology generation.

show abstract

Accelerating business analytics applications

Cited by 26 publications

References 32 publications

Navigating big data with high-throughput, energy-efficient data partitioning

Navigating big data with high-throughput, energy-efficient data partitioning

Q100

Energy Analysis of Hardware and Software Range Partitioning

Contact Info

Product

Resources

About