arXiv subjects

Anastasios Gounaris

Publications and source records attributed to Anastasios Gounaris.

18 recordsLinked to original sources

arXiv2025

Unfolding Data Quality Dimensions in Practice: A Survey

Data quality describes the degree to which data meet specific requirements and are fit for use by humans and/or downstream tasks (e.g., artificial intelligence). Data quality can be assessed across multiple high-level concepts called dimensions, such as accuracy, completeness, consistency, or timeliness. While extensive research and several attempts for standardization (e.g., ISO/IEC 25012) exist for data quality dimensions, their practical application often remains unclear. In parallel to research endeavors, a large number of tools have been developed that implement functionalities for the detection and mitigation of specific data quality issues, such as missing values or outliers. With this paper, we aim to bridge this gap between data quality theory and practice by systematically connecting low-level functionalities offered by data quality tools with high-level dimensions, revealing their many-to-many relationships. Through an examination of seven open-source data quality tools, we provide a comprehensive mapping between their functionalities and the data quality dimensions, demonstrating how individual functionalities and their variants partially contribute to the assessment of single dimensions. This systematic survey provides both practitioners and researchers with a unified view on the fragmented landscape of data quality checks, offering actionable insights for quality assessment across multiple dimensions.

cs.DB↗

arXiv2025

Handling out-of-order input arrival in CEP engines on the edge combining optimistic, pessimistic and lazy evaluation

In Complex Event Processing, handling out-of-order, late, and duplicate events is critical for real-time analytics, especially on resource-constrained devices that process heterogeneous data from multiple sources. We present LimeCEP, a hybrid CEP approach that combines lazy evaluation, buffering, and speculative processing to efficiently handle data inconsistencies while supporting multi-pattern detection under relaxed semantics. LimeCEP integrates Kafka for efficient message ordering, retention, and duplicate elimination, and offers configurable strategies to trade off between accuracy, latency, and resource consumption. Compared to state-of-the-art systems like SASE and FlinkCEP, LimeCEP achieves up to six orders of magnitude lower latency, with up to 10 times lower memory usage and 6 times lower CPU utilization, while maintaining near-perfect precision and recall under high-disorder input streams, making it well-suited for non-cloud deployments.

cs.DB↗

arXiv2025

Stream DaQ: Stream-First Data Quality Monitoring

Data quality is fundamental to modern data science workflows, where data continuously flows as unbounded streams feeding critical downstream tasks, from elementary analytics to advanced artificial intelligence models. Existing data quality approaches either focus exclusively on static data or treat streaming as an extension of batch processing, lacking the temporal granularity and contextual awareness required for true streaming applications. In this paper, we present a novel data quality monitoring model specifically designed for unbounded data streams. Our model introduces stream-first concepts, such as configurable windowing mechanisms, dynamic constraint adaptation, and continuous assessment that produces quality meta-streams for real-time pipeline awareness. To demonstrate practical applicability, we developed Stream DaQ, an open-source Python framework that implements our theoretical model. Stream DaQ unifies and adapts over 30 quality checks fragmented across existing static tools into a comprehensive streaming suite, enabling practitioners to define sophisticated, context-aware quality constraints through compositional expressiveness. Our evaluation demonstrates that the model's implementation significantly outperforms a production-grade alternative in both execution time and throughput while offering richer functionality via native streaming capabilities compared to other choices. Through its Python-native design, Stream DaQ seamlessly integrates with modern data science workflows, making continuous quality monitoring accessible to the broader data science community.

cs.DB↗

arXiv2025

Storing and Querying Evolving Graphs in NoSQL Storage Models

This paper investigates advanced storage models for evolving graphs, focusing on the efficient management of historical data and the optimization of global query performance. Evolving graphs, which represent dynamic relationships between entities over time, present unique challenges in preserving their complete history while supporting complex analytical queries. We first do a fast review of the current state of the art focusing mainly on distributed historical graph databases to provide the context of our proposals. We investigate the im- plementation of an enhanced vertex-centric storage model in MongoDB that prioritizes space efficiency by leveraging in-database query mechanisms to minimize redundant data and reduce storage costs. To ensure broad applicability, we employ datasets, some of which are generated with the LDBC SNB generator, appropriately post-processed to utilize both snapshot- and interval-based representations. Our experimental results both in centralized and distributed infrastructures, demonstrate significant improvements in query performance, particularly for resource-intensive global queries that traditionally suffer from inefficiencies in entity-centric frameworks. The proposed model achieves these gains by optimizing memory usage, reducing client involvement, and exploiting the computational capabilities of MongoDB. By addressing key bottlenecks in the storage and processing of evolving graphs, this study demonstrates a step toward a robust and scalable framework for managing dynamic graph data. This work contributes to the growing field of temporal graph analytics by enabling more efficient ex- ploration of historical data and facilitating real-time insights into the evolution of complex networks.

cs.DB↗

arXiv2024

A survey of open-source data quality tools: shedding light on the materialization of data quality dimensions in practice

Data Quality (DQ) describes the degree to which data characteristics meet requirements and are fit for use by humans and/or systems. There are several aspects in which DQ can be measured, called DQ dimensions (i.e. accuracy, completeness, consistency, etc.), also referred to as characteristics in literature. ISO/IEC 25012 Standard defines a data quality model with fifteen such dimensions, setting the requirements a data product should meet. In this short report, we aim to bridge the gap between lower-level functionalities offered by DQ tools and higher-level dimensions in a systematic manner, revealing the many-to-many relationships between them. To this end, we examine 6 open-source DQ tools and we emphasize on providing a mapping between the functionalities they offer and the DQ dimensions, as defined by the ISO standard. Wherever applicable, we also provide insights into the software engineering details that tools leverage, in order to address DQ challenges.

cs.DB↗

arXiv2024

A Comprehensive Scalable Framework for Cloud-Native Pattern Detection with Enhanced Expressiveness

Detecting complex patterns in large volumes of event logs has diverse applications in various domains, such as business processes and fraud detection. Existing systems like ELK are commonly used to tackle this challenge, but their performance deteriorates for large patterns, while they suffer from limitations in terms of expressiveness and explanatory capabilities for their responses. In this work, we propose a solution that integrates a Complex Event Processing (CEP) engine into a broader query processsor on top of a decoupled storage infrastructure containing inverted indices of log events. The results demonstrate that our system excels in scalability and robustness, particularly in handling complex queries. Notably, our proposed system delivers responses for large complex patterns within seconds, while ELK experiences timeouts after 10 minutes. It also significantly outperforms solutions relying on FlinkCEP and executing MATCH_RECOGNIZE SQL queries.

cs.DB↗

arXiv2022

Rank-based Heuristics for Optimizing the Execution of Product Data Models

The Product Data Model (PDM) is an example of a data-centric approach to modelling information-intensive business processes, which offers exibility and facilitates process optimization. Because the approach is declarative in nature, there may be multiple, alternative execution plans that can produce the desired end product. To generate such plans, several heuristics have been proposed in the literature. The contributions of this work are twofold: (i) we propose new heuristics that capitalize on established techniques for optimizing data-intensive work ows in terms of execution time and cost and transfer them to business processes; and (ii) we extensively evaluate the existing solutions. Our results shed light on the merits of each heuristic and show that our new heuristics can yield significant benefits.

cs.DB↗

arXiv2021

Cost models for geo-distributed massively parallel streaming analytics

This report is part of the DataflowOpt project on optimization of modern dataflows and aims to introduce a data quality-aware cost model that covers the following aspects in combination: (1) heterogeneity in compute nodes, (2) geo-distribution, (3) massive parallelism, (4) complex DAGs and (5) streaming applications. Such a cost model can be then leveraged to devise cost-based optimization solutions that deal with task placement and operator configuration.

cs.DB↗

arXiv2021

Analysis of key flavors of event-driven predictive maintenance using logs of phenomena described by Weibull distributions

This work explores two approaches to event-driven predictive maintenance in Industry 4.0 that cast the problem at hand as a classification or a regression one, respectively, using as a starting point two state-of-the-art solutions. For each of the two approaches, we examine different data preprocessing techniques, different prediction algorithms and the impact of ensemble and sampling methods. Through systematic experiments regarding the aspectsmentioned above,we aimto understand the strengths of the alternatives, and more importantly, shed light on how to navigate through the vast number of such alternatives in an informed manner. Our work constitutes a key step towards understanding the true potential of this type of data-driven predictive maintenance as of to date, and assist practitioners in focusing on the aspects that have the greatest impact.

cs.LG↗

arXiv2019

Adaptive filter ordering in Spark

This report describes a technical methodology to render the Apache Spark execution engine adaptive. It presents the engineering solutions, which specifically target to adaptively reorder predicates in data streams with evolving statistics. The system extension developed is available as an open-source prototype. Indicative experimental results show its overhead and sensitivity to tuning parameters.

cs.DB↗

arXiv2019

Continuous Outlier Mining of Streaming Data in Flink

In this work, we focus on distance-based outliers in a metric space, where the status of an entity as to whether it is an outlier is based on the number of other entities in its neighborhood. In recent years, several solutions have tackled the problem of distance-based outliers in data streams, where outliers must be mined continuously as new elements become available. An interesting research problem is to combine the streaming environment with massively parallel systems to provide scalable streambased algorithms. However, none of the previously proposed techniques refer to a massively parallel setting. Our proposal fills this gap and investigates the challenges in transferring state-of-the-art techniques to Apache Flink, a modern platform for intensive streaming analytics. We thoroughly present the technical challenges encountered and the alternatives that may be applied. We show speed-ups of up to 117 (resp. 2076) times over a naive parallel (resp. non-parallel) solution in Flink, by using just an ordinary four-core machine and a real-world dataset. When moving to a three-machine cluster, due to less contention, we manage to achieve both better scalability in terms of the window slide size and the data dimensionality, and even higher speed-ups, e.g., by a factor of 510. Overall, our results demonstrate that oulier mining can be achieved in an efficient and scalable manner. The resulting techniques have been made publicly available as open-source software.

cs.DB↗

arXiv2018

Speeding-up the Verification Phase of Set Similarity Joins in the GPGPU paradigm

We investigate the problem of exact set similarity joins using a co-process CPU-GPU scheme. The state-of-the-art CPU solutions split the wok in two main phases. First, filtering and index building takes place to reduce the candidate sets to be compared as much as possible; then the pairs are compared to verify whether they should become part of the result. We investigate in-depth solutions for transferring the second, so-called verification phase, to the GPU addressing several challenges regarding the data serialization and layout, the thread management and the techniques to compare sets of tokens. Using real datasets, we provide concrete experimental proofs that our solutions have reached their maximum potential, since they totally overlap verification with CPU tasks, and manage to yield significant speed-ups, up to 2.6X in our cases.

cs.DB↗

arXiv2017

The Many Faces of Data-centric Workflow Optimization: A Survey

Workflow technology is rapidly evolving and, rather than being limited to modeling the control flow in business processes, is becoming a key mechanism to perform advanced data management, such as big data analytics. This survey focuses on data-centric workflows (or workflows for data analytics or data flows), where a key aspect is data passing through and getting manipulated by a sequence of steps. The large volume and variety of data, the complexity of operations performed, and the long time such workflows take to compute give rise to the need for optimization. In general, data-centric workflow optimization is a technology in evolution. This survey focuses on techniques applicable to workflows comprising arbitrary types of data manipulation steps and semantic inter-dependencies between such steps. Further, it serves a twofold purpose. Firstly, to present the main dimensions of the relevant optimization problems and the types of optimizations that occur before flow execution. Secondly, to provide a concise overview of the existing approaches with a view to highlighting key observations and areas deserving more attention from the community.

cs.DB↗

arXiv2016

Spark Parameter Tuning via Trial-and-Error

Spark has been established as an attractive platform for big data analysis, since it manages to hide most of the complexities related to parallelism, fault tolerance and cluster setting from developers. However, this comes at the expense of having over 150 configurable parameters, the impact of which cannot be exhaustively examined due to the exponential amount of their combinations. The default values allow developers to quickly deploy their applications but leave the question as to whether performance can be improved open. In this work, we investigate the impact of the most important of the tunable Spark parameters on the application performance and guide developers on how to proceed to changes to the default values. We conduct a series of experiments with known benchmarks on the MareNostrum petascale supercomputer to test the performance sensitivity. More importantly, we offer a trial-and-error methodology for tuning parameters in arbitrary applications based on evidence from a very small number of experimental runs. We test our methodology in three case studies, where we manage to achieve speedups of more than 10 times.

cs.DC↗

arXiv2015

Towards Automated Performance Optimization of BPMN Business Processes

Business Process Model and Notation (BPMN) provides a standard for the design of business processes. It focuses on bridging the gap between the analysis and the technical perspectives, and aims to deliver process automation. The aim of this technical report is to complement this effort by transferring knowledge from the related field of data-centric workflows aiming to provide automated performance optimization of the business process execution. Automated optimization lifts a burden from BPMN designers and increases workflow flexibility and resilience. As a key step towards this goal, the contribution of this work is to provide a methodology to map BPMNv2.0 models to annotated directed acyclic graphs, which emphasize the volume of the tokens exchanged and are amenable to existing automated optimization algorithms. In addition, concrete examples of mappings are given, while the optimization opportunities that are opened are explained, thus providing insights into the potential performance benefits and we discuss technical research issues.

cs.DB↗

arXiv2015

Cost optimization of data flows based on task re-ordering

Analyzing big data in a highly dynamic environment becomes more and more critical because of the increasingly need for end-to-end processing of this data. Modern data flows are quite complex and there are not efficient, cost-based, fully-automated, scalable optimization solutions that can facilitate flow designers. The state-of-the-art proposals fail to provide near optimal solutions even for simple data flows. To tackle this problem, we introduce a set of approximate algorithms for defining the execution order of the constituent tasks, in order to minimize the total execution cost of a data flow. We also present the advantages of the parallel execution of data flows. We validated our proposals in both a real tool and synthetic flows and the results show that we can achieve significant speed-ups, moving much closer to optimal solutions.

cs.DB↗

arXiv2014

Cloud elasticity using probabilistic model checking

Cloud computing has become the leading paradigm for deploying large-scale infrastructures and running big data applications, due to its capacity of achieving economies of scale. In this work, we focus on one of the most prominent advantages of cloud computing, namely the on-demand resource provisioning, which is commonly referred to as elasticity. Although a lot of effort has been invested in developing systems and mechanisms that enable elasticity, the elasticity decision policies tend to be designed without guaranteeing or quantifying the quality of their operation. This work aims to make the development of elasticity policies more formalized and dependable. We make two distinct contributions. First, we propose an extensible approach to enforcing elasticity through the dynamic instantiation and online quantitative verification of Markov Decision Processes (MDP) using probabilistic model checking. Second, we propose concrete elasticity models and related elasticity policies. We evaluate our decision policies using both real and synthetic datasets in clusters of NoSQL databases. According to the experimental results, our approach improves upon the state-of-the-art in significantly increasing user-defined utility values and decreasing user-defined threshold violations.

cs.DC↗

arXiv2013

Skew Handling in Aggregate Streaming Queries on GPUs

Nowadays, the data to be processed by database systems has grown so large that any conventional, centralized technique is inadequate. At the same time, general purpose computation on GPU (GPGPU) recently has successfully drawn attention from the data management community due to its ability to achieve significant speed-ups at a small cost. Efficient skew handling is a well-known problem in parallel queries, independently of the execution environment. In this work, we investigate solutions to the problem of load imbalances in parallel aggregate queries on GPUs that are caused by skewed data. We present a generic load-balancing framework along with several instantiations, which we experimentally evaluate. To the best of our knowledge, this is the first attempt to present runtime load-balancing techniques for database operations on GPUs.

cs.DB↗