Taster: Self-Tuning, Elastic and Online Approximate Query Processing

Olma, Matthaios; Papapetrou, Odysseas; Appuswamy, Raja; Ailamaki, Anastasia

doi:10.1109/icde.2019.00050

Cited by 13 publications

(33 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, Taster [ 17 ] features the following main components: synopses collector and warehouse, cost-based planner, and tuner. The cost-based planner interactively generates a set of approximate execution plans.…”

Section: Related Workmentioning

confidence: 99%

QoS-Aware Approximate Query Processing for Smart Cities Spatial Data Streams

Jawarneh

Bellavista

Corradi

et al. 2021

Sensors

View full text Add to dashboard Cite

Large amounts of georeferenced data streams arrive daily to stream processing systems. This is attributable to the overabundance of affordable IoT devices. In addition, interested practitioners desire to exploit Internet of Things (IoT) data streams for strategic decision-making purposes. However, mobility data are highly skewed and their arrival rates fluctuate. This nature poses an extra challenge on data stream processing systems, which are required in order to achieve pre-specified latency and accuracy goals. In this paper, we propose ApproxSSPS, which is a system for approximate processing of geo-referenced mobility data, at scale with quality of service guarantees. We focus on stateful aggregations (e.g., means, counts) and top-N queries. ApproxSSPS features a controller that interactively learns the latency statistics and calculates proper sampling rates to meet latency or/and accuracy targets. An overarching trait of ApproxSSPS is its ability to strike a plausible balance between latency and accuracy targets. We evaluate ApproxSSPS on Apache Spark Structured Streaming with real mobility data. We also compared ApproxSSPS against a state-of-the-art online adaptive processing system. Our extensive experiments prove that ApproxSSPS can fulfill latency and accuracy targets with varying sets of parameter configurations and load intensities (i.e., transient peaks in data loads versus slow arriving streams). Moreover, our results show that ApproxSSPS outperforms the baseline counterpart by significant magnitudes. In short, ApproxSSPS is a novel spatial data stream processing system that can deliver real accurate results in a timely manner, by dynamically specifying the limits on data samples.

show abstract

Section: Related Workmentioning

confidence: 99%

QoS-Aware Approximate Query Processing for Smart Cities Spatial Data Streams

Jawarneh

Bellavista

Corradi

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…A highly desirable feature of any AQP engine is its ability to offer error guarantees to the user, regarding its approximated answers. For sampling-based AQP engines, providing such guarantees is relatively straightforward: Using subsampling/bootstraping methods, confidence intervals ([low, hiдh]) can be derived, associated with certain confidence levels (l%), indicating that the sampled-population statistic of interest (say, the mean) will fall within the [low, hiдh] range with probability l% [9,31,44,47]. However, this is not appropriate for AQP engines which, instead of inferring population statistics of a sampled population, employ predictive-ML-based models for predicting answers to future questions.…”

Section: Error Guaranteesmentioning

confidence: 99%

“…Their results have to be computed over large quantities of data. Therefore, research in AQP has been pretty strong the last decades [7,9,14,16,24,27,30,31,41,44,[47][48][49]64] and still the list is not exhaustive. We can categorize most AQP engines to based AQP engines [9,44,47] and online aggregation engines [27,64].…”

Section: Related Workmentioning

confidence: 99%

“…Therefore, research in AQP has been pretty strong the last decades [7,9,14,16,24,27,30,31,41,44,[47][48][49]64] and still the list is not exhaustive. We can categorize most AQP engines to based AQP engines [9,44,47] and online aggregation engines [27,64]. Sampling based AQP engines create samples over some (or all) of the columns in tables and produce answers with error guarantees based on samples.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

ML-AQP: Query-Driven Approximate Query Processing based on Machine Learning

Savva¹,

Anagnostopoulos²,

Triantafillou³

2020

Preprint

View full text Add to dashboard Cite

As more and more organizations rely on data-driven decision making, large-scale analytics become increasingly important. However, an analyst is often stuck waiting for an exact result. As such, organizations turn to Cloud providers that have infrastructure for efficiently analyzing large quantities of data. But, with increasing costs, organizations have to optimize their usage. Having a cheap alternative that provides speed and efficiency will go a long way. Concretely, we offer a solution that can provide approximate answers to aggregate queries, relying on Machine Learning (ML), which is able to work alongside Cloud systems. Our developed lightweight ML-led system can be stored on an analyst's local machine or deployed as a service to instantly answer analytic queries, having low response times and monetary/computational costs and energy footprint. To accomplish this we leverage the knowledge obtained by previously answered queries and build ML models that can estimate the result of new queries in an efficient and inexpensive manner. The capabilities of our system are demonstrated using extensive evaluation with both real and synthetic datasets/workloads and well known benchmarks.

show abstract

“…Recently, some methods make efforts to adopt machine learning methods to solve the AQP problems [23,26]. The main drawback of these methods is that the query estimation based on the learned model cannot provide a priori error guarantee as the sampling-based methods [23,26,31]. However, the error guarantee is an important metric to measure the quality of the AQP estimations [2].…”

Section: Introductionmentioning

confidence: 99%

LAQP: Learning-based Approximate Query Processing

Zhang

Wang

2020

Preprint

View full text Add to dashboard Cite

Querying on big data is a challenging task due to the rapid growth of data amount. Approximate query processing (AQP) is a way to meet the requirement of fast response. In this paper, we propose a learning-based AQP method called the LAQP. The LAQP builds an error model learned from the historical queries to predict the sampling-based estimation error of each new query. It makes a combination of the sampling-based AQP, the pre-computed aggregations and the learned error model to provide high-accurate query estimations with a small off-line sample. The experimental results indicate that our LAQP outperforms the sampling-based AQP, the pre-aggregation-based AQP and the most recent learning-based AQP method.

show abstract

Taster: Self-Tuning, Elastic and Online Approximate Query Processing

Cited by 13 publications

References 39 publications

QoS-Aware Approximate Query Processing for Smart Cities Spatial Data Streams

QoS-Aware Approximate Query Processing for Smart Cities Spatial Data Streams

ML-AQP: Query-Driven Approximate Query Processing based on Machine Learning

LAQP: Learning-based Approximate Query Processing

Contact Info

Product

Resources

About