Approximate Query Processing Research Articles

Learning-Based Sample Tuning for Approximate Query Processing in Interactive Data Exploration

Learning-Based Sample Tuning for Approximate Query Processing in Interactive Data Exploration

Learned Optimizer for Online Approximate Query Processing in Data Exploration

Learned Optimizer for Online Approximate Query Processing in Data Exploration

Machine Unlearning in Learned Databases: An Experimental Analysis

Machine learning models based on neural networks (NNs) are enjoying ever-increasing attention in the Database (DB) community, both in research and practice. However, an important issue has been largely overlooked, namely the challenge of dealing with the inherent, highly dynamic nature of DBs, where data updates are fundamental, highly-frequent operations (unlike, for instance, in ML classification tasks). Although some recent research has addressed the issues of maintaining updated NN models in the presence of new data insertions, the effects of data deletions (a.k.a., "machine unlearning") remain a blind spot. With this work, for the first time to our knowledge, we pose and answer the following key questions: What is the effect of unlearning algorithms on NN-based DB models? How do these effects translate to effects on key downstream DB tasks, such as cardinality/selectivity estimation (SE), approximate query processing (AQP), data generation (DG), and upstream tasks like data classification (DC)? What metrics should we use to assess the impact and efficacy of unlearning algorithms in learned DBs? Is the problem of (and solutions for) machine unlearning in DBs different from that of machine learning in DBs in the face of data insertions? Is the problem of (and solutions for) machine unlearning for DBs different from unlearning in the ML literature? what are the overhead and efficiency of unlearning algorithms (versus the naive solution of retraining from scratch)? What is the sensitivity of unlearning on batching delete operations (in order to reduce model updating overheads)? If we have a suitable unlearning algorithm (forgetting old knowledge), can we combine it with an algorithm handling data insertions (new knowledge) en route to solving the general adaptability/updatability requirement in learned DBs in the face of both data inserts and deletes? We answer these questions using a comprehensive set of experiments, various unlearning algorithms, a variety of downstream DB tasks (such as SE, AQP, and DG), and an upstream task (DC), each with different NNs, and using a variety of metrics (model-internal, and downstream-task specific) on a variety of real datasets, making this also a first key step towards a benchmark for learned DB unlearning.

Forester: Approximate Processing of an Imperative Procedure for Query-Time Exploratory Data Analysis in a Relational Database

Query-time Exploratory Data Analysis (qEDA) is an increasingly demanding aspect of the data analysis process that entails visually and quantitatively summarizing, comprehending, and interpreting the primary characteristics of a dataset. Nowadays, an imperative procedure is popular in relational databases for EDA because it enables us to write multiple dependent declarative queries with imperative logic. As online analytical processing (OLAP) systems contain extremely large datasets, data scientists often need quick visualizations of data, using approximate processing of imperative procedures, before analyzing them in their entirety. We identify gaps in the existing techniques, in that they are unable to sample both declarative-dependent statements and control logic at the same time and perform multi-dependent sampling-based approximate processing within the permitted time in qEDA. Traditional approximate query processing (AQP) involves tuple sampling for a single query approximation and enables queries to be executed over arbitrary random samples of tables. However, available AQP methods cannot produce a further representative sample of the data distribution for the dependent statements to estimate accurately and quickly for multiple dependent statements. On the other hand, sampling control structures, like loops and conditional statements, are discussed separately, without regard to the imperative structure of statements in a procedure. In this study, we propose Forester, a novel agile approximate processing method for imperative procedures that performs imperative program-aware sampling, which includes both statements with control regions (i.e., branch and loop) and processes them approximately within the permitted time in qEDA. Our method produces more targeted samples for each relation, while maintaining the data and control flow of dependent queries and imperative logic and determining all the conditions for a relation across all the statements in the sample that guarantee the existence of relevant data for dependent data distribution. Utilizing a workload of multi-statement imperative procedures from the Transaction Processing Performance Council Decision Support (TPC-DS) database, our experiment demonstrates that Forester outperforms the existing system in sampling, producing minimum error, and improving response time.

PairwiseHist: Fast, Accurate and Space-Efficient Approximate Query Processing with Data Compression

Exponential growth in data collection is creating significant challenges for data storage and analytics latency. Approximate Query Processing (AQP) has long been touted as a solution for accelerating analytics on large datasets, however, there is still room for improvement across all key performance criteria. In this paper, we propose a novel histogram-based data synopsis called PairwiseHist that uses recursive hypothesis testing to ensure accurate histograms and can be built on top of data compressed using Generalized Deduplication (GD). We thus show that GD data compression can contribute to AQP. Compared to state-of-the-art AQP approaches, Pairwise-Hist achieves better performance across all key metrics, including 2.6× higher accuracy, 3.5× lower latency, 24× smaller synopses and 1.5--4× faster construction time.

GAN-Based Tabular Data Generator for Constructing Synopsis in Approximate Query Processing: Challenges and Solutions

In data-driven systems, data exploration is imperative for making real-time decisions. However, big data are stored in massive databases that are difficult to retrieve. Approximate Query Processing (AQP) is a technique for providing approximate answers to aggregate queries based on a summary of the data (synopsis) that closely replicates the behavior of the actual data; this can be useful when an approximate answer to queries is acceptable in a fraction of the real execution time. This study explores the novel utilization of a Generative Adversarial Network (GAN) for the generation of tabular data that can be employed in AQP for synopsis construction. We thoroughly investigate the unique challenges posed by the synopsis construction process, including maintaining data distribution characteristics, handling bounded continuous and categorical data, and preserving semantic relationships, and we then introduce the advancement of tabular GAN architectures that overcome these challenges. Furthermore, we propose and validate a suite of statistical metrics tailored for assessing the reliability of GAN-generated synopses. Our findings demonstrate that advanced GAN variations exhibit a promising capacity to generate high-fidelity synopses, potentially transforming the efficiency and effectiveness of AQP in data-driven systems.

Guaranteeing an Exact Error Bound for Bounded Approximate Query Processing

Guaranteeing an Exact Error Bound for Bounded Approximate Query Processing

Secure Sampling for Approximate Multi-party Query Processing

We study the problem of random sampling in the secure multi-party computation (MPC) model. In MPC, taking a sample securely must have a cost Ω(n) irrespective to the sample size s. This is in stark contrast with the plaintext setting, where a sample can be taken in O(s) time trivially. Thus, the goal of approximate query processing (AQP) with sublinear costs seems unachievable under MPC. To get around this inherent barrier, in this paper we take a two-stage approach: In the offline stage, we generate a batch of n/s samples with (n) total cost, which can then be consumed to answer queries as they arrive online. Such an approach allows us to achieve an Õ(s) amortized cost per query, similar to the plaintext setting. Based on our secure batch sampling algorithms, we build MASQUE, an MPC-AQP system that achieves sublinear online query costs by running an MPC protocol to evaluate the queries on pre-generated samples. MASQUE achieves the strong security guarantee of the MPC model, i.e., nothing is revealed beyond the query result, which itself can be further protected by (amplified) differential privacy

Polygon Simplification for the Efficient Approximate Analytics of Georeferenced Big Data.

The unprecedented availability of sensor networks and GPS-enabled devices has caused the accumulation of voluminous georeferenced data streams. These data streams offer an opportunity to derive valuable insights and facilitate decision making for urban planning. However, processing and managing such data is challenging, given the size and multidimensionality of these data. Therefore, there is a growing interest in spatial approximate query processing depending on stratified-like sampling methods. However, in these solutions, as the number of strata increases, response time grows, thus counteracting the benefits of sampling. In this paper, we originally show the design and realization of a novel online geospatial approximate processing solution called GeoRAP. GeoRAP employs a front-stage filter based on the Ramer-Douglas-Peucker line simplification algorithm to reduce the size of study area coverage; thereafter, it employs a spatial stratified-like sampling method that minimizes the number of strata, thus increasing throughput and minimizing response time, while keeping the accuracy loss in check. Our method is applicable for various online and batch geospatial processing workloads, including complex geo-statistics, aggregation queries, and the generation of region-based aggregate geo-maps such as choropleth maps and heatmaps. We have extensively tested the performance of our prototyped solution with real-world big spatial data, and this paper shows that GeoRAP can outperform state-of-the-art baselines by an order of magnitude in terms of throughput while statistically obtaining results with good accuracy.

ShadowAQP: Efficient Approximate Group-by and Join Query via Attribute-Oriented Sample Size Allocation and Data Generation

Approximate query processing (AQP) is one of the key techniques to cope with big data querying problem on account that it obtains approximate answers efficiently. To address non-trivial sample selection and heavy sampling cost issues in AQP, we propose ShadowAQP, an efficient and accurate approach based on attribute-oriented sample size allocation and data generation. We select samples according to group-by and join attributes, and determine the sample size for each group of unique value combinations to improve query accuracy. We design a conditional variational autoencoder model with automatic table data encoding and model update strategies. To further improve accuracy and efficiency, we propose a set of extensions, including parallel multi-round sampling aggregation, data outlier-aware sampling, and dimension reduction optimization. Evaluation results on diversified datasets show that, compared with SOTA approaches, ShadowAQP achieves 5.8× query speed performance improvement on average (up to 12.8×), while reducing query error by 74% on average (up to 95%) at the same time.

Approximate Queries over Concurrent Updates

Approximate Query Processing (AQP) systems produce estimation of query answers using small random samples. It is attractive for the users who are willing to trade accuracy for low query latency. On the other hand, real-world data are often subject to concurrent updates. If the user wants to perform real-time approximate data analysis, the AQP system must support concurrent updates and sampling. Towards that, we recently developed a new concurrent index, AB-tree, to support efficient sampling under updates. In this work, we will demonstrate the feasibility of supporting realtime approximate data analysis in online transaction settings using index-assisted sampling.

LAQy: Efficient and Reusable Query Approximations via Lazy Sampling

Modern analytical engines rely on Approximate Query Processing (AQP) to provide faster response times than the hardware allows for exact query answering. However, existing AQP methods impose steep performance penalties as workload unpredictability increases. Specifically, offline AQP relies on predictable workloads to create samples that match the queries in a priori to query execution, reducing query response times when queries match the expected workload. As soon as workload predictability diminishes, existing online AQP methods create query-specific samples with little reuse across queries, producing significantly smaller gains in response times. As a result, existing approaches cannot fully exploit the benefits of sampling under increased unpredictability. We analyze sample creation and propose LAQy, a framework for building, expanding, and merging samples to adapt to the changes in workload predicates. We show the main parameters that affect the sample creation time and propose lazy sampling to overcome the unpredictability issues that cause fast-but-specialized samples to be query-specific. We evaluate LAQy by implementing it in an in-memory code-generation-based scale-up analytical engine to show the adaptivity and practicality of our framework in a modern system. LAQy speeds up online sampling processing as a function of sample reuse ranging from practically zero to full online sampling time and from 2.5x to 19.3x in a simulated exploratory workload.

Detect, Distill and Update: Learned DB Systems Facing Out of Distribution Data

Machine Learning (ML) is changing DBs as many DB components are being replaced by ML models. One open problem in this setting is how to update such ML models in the presence of data updates. We start this investigation focusing on data insertions (dominating updates in analytical DBs). We study how to update neural network (NN) models when new data follows a different distribution (a.k.a. it is "out-of-distribution" -- OOD), rendering previously-trained NNs inaccurate. A requirement in our problem setting is that learned DB components should ensure high accuracy for tasks on old and new data (e.g., for approximate query processing (AQP), cardinality estimation (CE), synthetic data generation (DG), etc.). This paper proposes a novel updatability framework (DDUp). DDUp can provide updatability for different learned DB system components, even based on different NNs, without the high costs to retrain the NNs from scratch. DDUp entails two components: First, a novel, efficient, and principled statistical-testing approach to detect OOD data. Second, a novel model updating approach, grounded on the principles of transfer learning with knowledge distillation, to update learned models efficiently, while still ensuring high accuracy. We develop and showcase DDUp's applicability for three different learned DB components, AQP, CE, and DG, each employing a different type of NN. Detailed experimental evaluation using real and benchmark datasets for AQP, CE, and DG detail DDUp's performance advantages.

Machine Learning based Approximate Query Processing for Women Health Analytics

Good health and well being is one of the most essential targets of the Sustainable Development Goals (SDGs). This paper primarily focuses on Preventive and Diagnostic care of Women Health because even today, women are disadvantaged by discrimination in many societies especially in rural sectors. Two main health issues, fetal abnormality in pregnant women and cervical cancer in women are analyzed so that the doctors and patients can be given early signals to take proactive measures. As per National Health Portal of India, around 1.7 million birth defects occur in India every year. So Antenatal Care(ANC) should be given utmost importance during Pregnancy in a woman's life. Cervical cancer is another issue prevalent amongst women, especialy over the age of 30. It's critical to catch it early and eliminate any risks that come with it. Here arises the need to develop a system that can analyze and predict the aforementioned anomalies at an early stage. This work proposes approximate query processing using different dynamic machine learning algorithms to analyze and predict the abnormalities. Further, a web application is built to facilitate the stakeholders, especially doctors, to interact with the system and iteratively query the system to understand the relationships amongst the various data variables and get appropriate predictions about fetal anomaly from CTG scans as well as presence of cervical cancer from various demographic information, habits, and historic medical records. Approximate Query Processing is accomplished using Correlation analysis, linear, and logistic regression algorithms in a dynamic, interactive, and iterative manner.

Conditional Generative Model Based Predicate-Aware Query Approximation

The goal of Approximate Query Processing (AQP) is to provide very fast but "accurate enough" results for costly aggregate queries thereby improving user experience in interactive exploration of large datasets. Recently proposed Machine-Learning-based AQP techniques can provide very low latency as query execution only involves model inference as compared to traditional query processing on database clusters. However, with increase in the number of filtering predicates (WHERE clauses), the approximation error significantly increases for these methods. Analysts often use queries with a large number of predicates for insights discovery. Thus, maintaining low approximation error is important to prevent analysts from drawing misleading conclusions. In this paper, we propose ELECTRA, a predicate-aware AQP system that can answer analytics-style queries with a large number of predicates with much smaller approximation errors. ELECTRA uses a conditional generative model that learns the conditional distribution of the data and at run-time generates a small (≈ 1000 rows) but representative sample, on which the query is executed to compute the approximate result. Our evaluations with four different baselines on three real-world datasets show that ELECTRA provides lower AQP error for large number of predicates compared to baselines.

Comparison of Data Sampling Strategies for Approximate Processing of Queries to a Large Database

Interactive analytics requires database systems to promptly process aggregation queries. As the volume of data continues to grow at a high rate, this becomes more and more challenging. In the past, the database community has come up with two ideas to solve this problem: Approximate Query Processing (AQP) and Aggregate Precomputation (AggPre). In this article, we will focus on APQ. Many techniques were proposed for optimizing AQP performance. Most of them are based on the creation of the best stratified samples. But none of the existing approaches can answer the following questions: "Under what conditions will the sampling strategy be optimal? What is the maximum error in estimating an aggregate when using the strategy in case of data skew?" The article provides answers to the questions posed. The article discusses three strategies for selecting segments from a data array: 1) uniform, 2) based on the distribution of read records by segments, 3) based on the distribution of read records and aggregate values by segments. The article defines conditions for the optimality of each strategy. Proposed analytical methods solve the optimization problem and determine the conditions under which the value of the confidence interval of the estimated aggregate will be minimal. The paper also studied the effect of data skews on the accuracy of the estimation of aggregated values. The maximum value of the confidence interval is calculated at the border of the admissible region. Experiments on depersonalized banMng transactions demonstrated that the 3rd sampling strategy is much better than the first two strategies.

AB-tree

There has been an increasing demand for real-time data analytics. Approximate Query Processing (AQP) is a popular option for that because it can use random sampling to trade some accuracy for lower query latency. However, the state-of-the-art AQP system either relies on scan-based sampling algorithms to draw samples, which can still incur a non-trivial cost of table scan, or creates samples of the database in a preprocessing step, which are hard to update. The alternative is to use the aggregate B-tree indexes to support both random sampling and updates in database with logarithmic time. However, to the best of our knowledge, it is unknown how to design an aggregate B-tree to support highly concurrent random sampling and updates, due to the difficulty of maintaining the aggregate weights correctly and efficiently with concurrency. In this work, we identify the key challenges to achieve high concurrency and present AB-tree, an index for highly concurrent random sampling and update operations. We also conduct extensive experiments to show its efficiency and efficacy in a variety of workloads.

Multi-relation Graph Summarization

Graph summarization is beneficial in a wide range of applications, such as visualization, interactive and exploratory analysis, approximate query processing, reducing the on-disk storage footprint, and graph processing in modern hardware. However, the bulk of the literature on graph summarization surprisingly overlooks the possibility of having edges of different types. In this article, we study the novel problem of producing summaries of multi-relation networks, i.e., graphs where multiple edges of different types may exist between any pair of nodes. Multi-relation graphs are an expressive model of real-world activities, in which a relation can be a topic in social networks, an interaction type in genetic networks, or a snapshot in temporal graphs. The first approach that we consider for multi-relation graph summarization is a two-step method based on summarizing each relation in isolation, and then aggregating the resulting summaries in some clever way to produce a final unique summary. In doing this, as a side contribution, we provide the first polynomial-time approximation algorithm based on the k -Median clustering for the classic problem of lossless single-relation graph summarization. Then, we demonstrate the shortcomings of these two-step methods, and propose holistic approaches, both approximate and heuristic algorithms, to compute a summary directly for multi-relation graphs. In particular, we prove that the approximation bound of k -Median clustering for the single relation solution can be maintained in a multi-relation graph with proper aggregation operation over adjacency matrices corresponding to its multiple relations. Experimental results and case studies (on co-authorship networks and brain networks) validate the effectiveness and efficiency of the proposed algorithms.

MISS: finding optimal sample sizes for approximate analytics

Nowadays, sampling-based Approximate Query Processing (AQP) is widely regarded as a promising way to achieve interactivity in big data analytics. To build such an AQP system, finding the minimal sample size for a query regarding given error constraints in general, called Sample Size Optimization (SSO), is an essential yet unsolved problem. Ideally, the goal of solving the SSO problem is to achieve statistical accuracy, computational efficiency and broad applicability all at the same time. Existing approaches either make idealistic assumptions on the statistical properties of the query, or completely disregard them. This may result in overemphasizing only one of the three goals while neglect the others. To overcome these limitations, we first examine carefully the statistical properties shared by common analytical queries. Then, based on the properties, we propose a linear model describing the relationship between sample sizes and the approximation errors of a query, which is called the error model. Then, we propose a Model-guided Iterative Sample Selection (MISS) framework to solve the SSO problem generally. Afterwards, based on the MISS framework, we propose a concrete algorithm, called $$L^{2}\textsc{Miss}$$ , to find optimal sample sizes under the $$L^{2}$$ norm error metric. Moreover, we extend the $$L^{2}\textsc{Miss}$$ algorithm to handle other error metrics. Finally, we show theoretically and empirically that the $$L^{2}\textsc{Miss}$$ algorithm and its extensions achieve satisfactory accuracy and efficiency for a considerably wide range of analytical queries.

Davos

Recently, a new horizon in data analytics, prescriptive analytics, is becoming more and more important to make data-driven decisions. As opposed to the progress of democratizing data acquisition and access, making data-driven decisions remains a significant challenge for people without technical expertise. In this regard, existing tools for data analytics which were designed decades ago still present a high bar for domain experts, and removing this bar requires a fundamental rethinking of both interface and backend. At Einblick, an MIT/Brown spin-off based on the Northstar project, we have been building the next generation analytics tool in the last few years. To overcome the shortcomings of existing processing engines, we propose Davos , Einblick's novel backend. Davos combines aspects of progressive computation, approximate query processing and sampling, with a specific focus on supporting user-defined operations. Moreover, Davos optimizes multi-tenant scenarios to promote collaboration. Both empirical evaluation and user study verify that Davos can greatly empower data analytics for new needs.