WIQ: Work-Intensive Query Scheduling for In-Memory Database Systems

Kraft, Stephan; Casale, Giuliano; Jula, Alin; Kilpatrick, Peter; Greer, Des

doi:10.1109/cloud.2012.120

Cited by 12 publications

(20 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In fact, we are not only interested into mean performance estimates but also want to consider the more expressive probabilistic measures. The difficulty of using analytical methods to obtain mean estimates from fork-join models that consider variable threading levels is discussed in [14], [16], [17]. However, obtaining probabilistic estimates for such models adds additional layers of complexity.…”

Section: B Predictive Functionalitymentioning

confidence: 99%

“…To further emphasize the importance of class-switching we compared our simulation model against the simulator in [14] that does not consider such a concept. The advantage of our model is that we can define the total workload population N as the maximum number of TPC-H clients that submit query requests.…”

Section: A Model Descriptionmentioning

confidence: 99%

“…The advantage of our model is that we can define the total workload population N as the maximum number of TPC-H clients that submit query requests. For the simulator in [14] however, we have to oversize the maximum workload population and set N for each query class separately to the maximum number of TPC-H clients. As shown in [14] this approach requires think-time estimates from trace logs dependent on the number of active clients, from which our class-switching model is independent.…”

Section: A Model Descriptionmentioning

confidence: 99%

“…For the simulator in [14] however, we have to oversize the maximum workload population and set N for each query class separately to the maximum number of TPC-H clients. As shown in [14] this approach requires think-time estimates from trace logs dependent on the number of active clients, from which our class-switching model is independent. To analyze the accuracy of both simulation models, we compared their response time predictions against trace response times for the 16 user scenario of our traces in [11].…”

Section: A Model Descriptionmentioning

confidence: 99%

See 3 more Smart Citations

Experiments or simulation? A characterization of evaluation methods for in-memory databases

Molka

Casale

2015

2015 11th International Conference on Network and Service Management (CNSM)

Self Cite

View full text Add to dashboard Cite

Abstract-The recent growth of interest for in-memory databases poses the question on whether established prediction methods such as response surfaces and simulation are effective to describe the performance of these systems. In particular, the limited dependence of in-memory technologies on the disk makes methods such as simulation more appealing than in the past, since disks are difficult to simulate. To answer this question, we study an in-memory commercial solution, SAP HANA, deployed on a high-end server with 120 physical cores. First, we apply experimental design methods to generate response surfaces that describe database performance as a function of workload and hardware parameters. Next, we develop a class-switching queueing network model to predict in-memory database performance under similar scenarios. By comparing the applicability of the two approaches to modeling multi-tenancy, we find that both queueing and response surface models yield mean prediction errors in the range 5%-22% with respect to mean memory occupancy and response times, but the accuracy for the latter deteriorates in response surfaces as the number of experiments are reduced, whereas simulation is effective in all cases. This suggests that simulation can be very effective in performance prediction for in-memory database management.

show abstract

Section: B Predictive Functionalitymentioning

confidence: 99%

Section: A Model Descriptionmentioning

confidence: 99%

Section: A Model Descriptionmentioning

confidence: 99%

Section: A Model Descriptionmentioning

confidence: 99%

See 2 more Smart Citations

Experiments or simulation? A characterization of evaluation methods for in-memory databases

Molka

Casale

2015

2015 11th International Conference on Network and Service Management (CNSM)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Kafka combines off-line and on-line processing data to provide realtime computing and produce ad hoc solution. 6) SAP Hana SAP Hana [50] is Big data analytics platform which performs real-time analysis on predictive analysis, business processes and sentiment data processing. SAP HANA database is core of the real-time platform.…”

Section: ) Apache Kafkamentioning

confidence: 99%

Big Data Analytics: A Perspective View

Pandey¹

2017

IJARCSSE

View full text Add to dashboard Cite

Abstract--Big data is a term coined for massive data sets with more varied and complex structure and also having the difficulties of storing, analysing and visualizing it for further usages. The process of diving into large amounts of data to discover patterns and disguised correlations is named as big data analytics. These are certainly useful to companies or organizations by gaining richer and deeper insights to compete in market. Thus big data implementations need to be analysed and executed efficiently as possible. Big data analytics is methodology which automates the gathering, organizing, contextualizing, processing and analysing Big data i.e. large set of data to capture patterns help make better decisions. Big data analytics challenges the situation of the present infrastructure of data storage management and also statistical data estimation. This paper studies the content, scope, methods, advantages and challenges of big data and also discusses privacy issue concern on it. The motive of the proposed study is to provide better and significant insights from research prospects and also lays an overview of data analysis methodologies and tools which are currently being utilized or proposed in literature. This work will be quite useful for the future researchers in this domain and facilitate the development of optimal techniques to address Big data.

show abstract

The Role of Grid Technologies: A Next Level Combat with Big Data

Mishra

Patel²

2016

Studies in Big Data

View full text Add to dashboard Cite

WIQ: Work-Intensive Query Scheduling for In-Memory Database Systems

Cited by 12 publications

References 14 publications

Experiments or simulation? A characterization of evaluation methods for in-memory databases

Experiments or simulation? A characterization of evaluation methods for in-memory databases

Big Data Analytics: A Perspective View

The Role of Grid Technologies: A Next Level Combat with Big Data

Contact Info

Product

Resources

About