Query processing on smart SSDs

Do, Jaeyoung; Kee, Yang-Suk; Patel, Jignesh M.; Park, Chanik; Park, Kwanghyun; DeWitt, David J.

doi:10.1145/2463676.2465295

Cited by 179 publications

(22 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the storage device stores data using lossy algorithms, the system sacrifices support for exactcomputing. In-storage processing (ISP) General-purpose intelligent storage frameworks such as Willow [76], Samsung's SmartSSD [16], Morpheus [85], Biscuit [23], Summarizer [40] and FlashAbacus [96] allow applications to use the processing power inside SSDs. These platforms fall short of approximate computing for the following reasons.…”

Section: Alternative Approachesmentioning

confidence: 99%

Dynamic Multi-Resolution Data Storage

Lokhandwala

et al. 2019

Proceedings of the 52nd Annual IEEE/ACM International Symposium on Microarchitecture

View full text Add to dashboard Cite

Approximate computing that works on less precise data leads to significant performance gains and energy-cost reductions for compute kernels. However, without leveraging the full-stack design of computer systems, modern computer architectures undermine the potential of approximate computing. In this paper, we present Varifocal Storage, a dynamic multiresolution storage system that tackles challenges in performance, quality, flexibility and cost for computer systems supporting diverse application demands. Varifocal Storage dynamically adjusts the dataset resolution within a storage device, thereby mitigating the performance bottleneck of exchanging/preparing data for approximate compute kernels. Varifocal Storage introduces Autofocus and iFilter mechanisms to provide quality control inside the storage device and make programs more adaptive to diverse datasets. Varifocal Storage also offers flexible, efficient support for approximate and exact computing without exceeding the costs of conventional storage systems by (1) saving the raw dataset in the storage device, and (2) targeting operators that complement the power of existing SSD controllers to dynamically generate lower-resolution datasets. We evaluate the performance of Varifocal Storage by running applications on a heterogeneous computer with our prototype SSD. The results show that Varifocal Storage can speed up data resolution adjustments by 2.02× or 1.74× without programmer input. Compared to conventional approximate-computing architectures, Varifocal Storage speeds up the overall execution time by 1.52×.

show abstract

Section: Alternative Approachesmentioning

confidence: 99%

Dynamic Multi-Resolution Data Storage

Lokhandwala

et al. 2019

Proceedings of the 52nd Annual IEEE/ACM International Symposium on Microarchitecture

View full text Add to dashboard Cite

show abstract

“…Using solid state disks (SSDs) as a storage device by databases to improve energy-efficiency is another worthy direction. Studies such as [4] claim that using Smart SSD can be benefit from both the performance and the energy consumption perspectives.…”

Section: Storage Managementmentioning

confidence: 99%

Obre: Offer-Borrow-Reform-Evaluate Initiatives for Green DBMSS

Roukh

Bellatreche

Tziritas

2019

AzJHPC

View full text Add to dashboard Cite

In the last few years, we have been seeing a significant increase in research about the energy efficiency of hardware and software components by both academic and industry. Today, energy efficiency is one of the most challenging issues in the area of information technologies and communication. In data-centric applications, database management systems are one of the major energy consumers, in which, a large amount of data is queried by complex queries running daily. Designing and implementing of an energy-aware DBMS that enables significant energy conservation while processing queries become a necessary need. Traditionally, existing DBMSs focus to high-performance during query optimization phase, while totally ignoring the energy consumption of the queries. In this paper, we propose a methodology, supported by a tool called EcoProD, focusing on query optimizers. To show its effectiveness, we implement it in Post-greSQL DBMS aiming reducing energy consumption without degrading query response time. A mathematical cost model is used to estimate the energy consumption. Its parameters are identified by a machine learning technique. We conduct intensive experiments using our cost models and a measurement tool dedicated to compute energy using dataset of TPC-H benchmark. Based on the obtained results, a probabilistic proof to demonstrate the confidence bounds of our model and results is given. CCS Concepts: Information systems-Relational database model; DBMS engine architectures; Database query processing; Relational database query languages;

show abstract

“…Devices such as Smart SSDs [9,22,41] and Programmable SSDs [6] have shown promising results, but gains are often limited by the performance of the embedded processors in such power constrained devices. Embedding reconfigurable hardware in storage devices is being investigated as well.…”

Section: Related Workmentioning

confidence: 99%

BlueDBM

Jun

Liu

Lee

et al. 2015

Proceedings of the 42nd Annual International Symposium on Computer Architecture

116

View full text Add to dashboard Cite

Complex data queries, because of their need for random accesses, have proven to be slow unless all the data can be accommodated in DRAM. There are many domains, such as genomics, geological data and daily twitter feeds where the datasets of interest are 5TB to 20 TB. For such a dataset, one would need a cluster with 100 servers, each with 128GB to 256GBs of DRAM, to accommodate all the data in DRAM. On the other hand, such datasets could be stored easily in the flash memory of a rack-sized cluster. Flash storage has much better random access performance than hard disks, which makes it desirable for analytics workloads. In this paper we present BlueDBM, a new system architecture which has flashbased storage with in-store processing capability and a lowlatency high-throughput inter-controller network. We show that BlueDBM outperforms a flash-based system without these features by a factor of 10 for some important applications. While the performance of a ram-cloud system falls sharply even if only 5%~10% of the references are to the secondary storage, this sharp performance degradation is not an issue in BlueDBM. BlueDBM presents an attractive point in the cost-performance trade-off for Big Data analytics.

show abstract

Query processing on smart SSDs

Cited by 179 publications

References 18 publications

Dynamic Multi-Resolution Data Storage

Dynamic Multi-Resolution Data Storage

Obre: Offer-Borrow-Reform-Evaluate Initiatives for Green DBMSS

BlueDBM

Contact Info

Product

Resources

About