Efficient inference for graph neural networks (GNNs) on large knowledge graphs (KGs) is essential for many real-world applications. GNN inference queries are computationally expensive and vary in complexity, as each involves a different number of target nodes linked to subgraphs of diverse densities and structures. Existing acceleration methods, such as pruning, quantization, and knowledge distillation, instantiate smaller models but do not adapt them to the structure or semantics of individual queries. They also store models as monolithic files that must be fully loaded, and miss the opportunity to retrieve only the neighboring nodes and corresponding model components that are semantically relevant to the target nodes. These limitations lead to excessive data loading and redundant computation on large KGs. This paper presents KG-WISE, a task-driven inference paradigm for large KGs. KG-WISE decomposes trained GNN models into fine-grained components that can be partially loaded based on the structure of the queried subgraph. It employs large language models (LLMs) to generate reusable query templates that extract semantically relevant subgraphs for each task, enabling query-aware and compact model instantiation. We evaluate KG-WISE on six large KGs with up to 42 million nodes and 166 million edges. KG-WISE achieves up to 28x faster inference and 98% lower memory usage than state-of-the-art systems while maintaining or improving accuracy across both commercial and open-weight LLMs.

Dialogue benchmarks are crucial in training and evaluating chatbots engaging in domain-specific conversations. Knowledge graphs (KGs) represent semantically rich and well-organized data spanning various domains, such as DBLP, DBpedia, and YAGO. Traditionally, dialogue benchmarks have been manually created from documents, neglecting the potential of KGs in automating this process. Some question-answering benchmarks are automatically generated using extensive preprocessing from KGs, but they do not support dialogue generation. This paper introduces Chatty-Gen, a novel multi-stage retrieval-augmented generation platform for automatically generating high-quality dialogue benchmarks tailored to a specific domain using a KG. Chatty-Gen decomposes the generation process into manageable stages and uses assertion rules for automatic validation between stages. Our approach enables control over intermediate results to prevent time-consuming restarts due to hallucinations. It also reduces reliance on costly and more powerful commercial LLMs. Chatty-Gen eliminates upfront processing of the entire KG using efficient query-based retrieval to find representative subgraphs based on the dialogue context. Our experiments with several real and large KGs demonstrate that C hatty -G en significantly outperforms state-of-the-art systems and ensures consistent model and system performance across multiple LLMs of diverse capabilities, such as GPT-4o, Gemini 1.5, Llama 3, and Mistral.

Data-centric machine learning (ML) pipelines extend traditional ML pipelines—of feature transformations, hyper-parameter tuning, and model training—by additional pre-processing steps for data cleaning, data augmentation, and feature engineering to create high-quality data with good coverage. Finding effective data-centric ML pipelines is still a labor- and compute-intensive process though. While AutoML tools use effective search strategies, they struggle to scale with large datasets. Large language models (LLMs) show promise for code generation but face challenges in generating data-centric ML pipelines due to private datasets not seen during training, complex pre-processing requirements, and the need for mitigating hallucinations. These demands exceed typical code generation as it requires actions tailored to the characteristics and requirements of a particular dataset. This paper introduces CatDB, a comprehensive, LLM-based system for generating effective, error-free, and efficient data-centric ML pipelines. CatDB leverages data catalog information and refined metadata to dynamically create dataset-specific rules (instructions) to guide the LLM. Moreover, CatDB includes a robust mechanism for automatic validation and error handling of the generated pipeline. Our experimental results show that CatDB reliably generates effective ML pipelines across diverse datasets, achieving accuracy comparable to or better than existing LLM-based systems, standalone AutoML tools, and combined workflows of data cleaning and AutoML tools, while delivering up to orders of magnitude faster performance on large datasets.

A Knowledge Graph (KG) is a heterogeneous graph encompassing a diverse range of node and edge types. Heterogeneous Graph Neural Networks (HGNNs) are popular for training machine learning tasks like node classification and link prediction on KGs. However, HGNN methods exhibit excessive complexity influenced by the KG's size, density, and the number of node and edge types. AI practitioners handcraft a subgraph of a KG$G$relevant to a specific task. We refer to this subgraph as a task-oriented subgraph (TOSG), which contains a subset of task-related node and edge types in$G$. Training the task using TOSG instead of$G$alleviates the excessive computation required for a large KG. Crafting the TOSG demands a deep understanding of the KG's structure and the task's objectives. Hence, it is challenging and time-consuming. This paper proposes KG-TOSA, an approach to automate the TOSG extraction for task-oriented HGNN training on a large KG. In KG-TOSA, we define a generic graph pattern that captures the KG's local and global structure relevant to a specific task. We explore different techniques to extract subgraphs matching our graph pattern: namely (i) two techniques sampling around targeted nodes using biased random walk or influence scores, and (ii) a SPARQL-based extraction method leveraging RDF engines' built-in indices. Hence, it achieves negligible preprocessing overhead compared to the sampling techniques. We develop a benchmark of real KGs of large sizes and various tasks for node classification and link prediction. Our experiments show that KG-TOSA helps state-of-the-art HGNN methods reduce training time and memory usage by up to 70% while improving the model performance, e.g., accuracy and inference time.

In recent years, we have witnessed the growing interest from academia and industry in applying data science technologies to analyze large amounts of data. In this process, a myriad of artifacts (datasets, pipeline scripts, etc.) are created. However, there has been no systematic attempt to holistically collect and exploit all the knowledge and experiences that are implicitly contained in those artifacts. Instead, data scientists recover information and expertise from colleagues or learn via trial and error. Hence, this paper presents a scalable platform, KGLiDS, that employs machine learning and knowledge graph technologies to abstract and capture the semantics of data science artifacts and their connections. Based on this information, KGLiDS enables various downstream applications, such as data discovery and pipeline automation. Our comprehensive evaluation covers use cases in data discovery, data cleaning, transformation, and AutoML. It shows that KGLiDS is significantly faster with a lower memory footprint than the state-of-the-art systems while achieving comparable or better accuracy.

This vision paper proposes KGNet, an on-demand graph machine learning (GML) as a service on top of RDF engines to support GML-enabled SPARQL queries. KGNet automates the training of GML models on a KG by identifying a task-specific subgraph. This helps reduce the task-irrelevant KG structure and properties for better scalability and accuracy. While training a GML model on KG, KGNet collects metadata of trained models in the form of an RDF graph called KGMeta, which is interlinked with the relevant subgraphs in KG. Finally, all trained models are accessible via a SPARQL-like query. We call it a GML-enabled query and refer to it as SPARQLML. KGNet supports SPARQLMLon top of existing RDF engines as an interface for querying and inferencing over KGs using GML models. The development of KGNet poses research opportunities in several areas, including meta-sampling for identifying task-specific subgraphs, GML pipeline automation with computational constraints, such as limited time and memory budget, and SPARQLMLquery optimization. KGNet supports different GML tasks, such as node classification, link prediction, and semantic entity matching. We evaluated KGNet using two real KGs of different application domains. Compared to training on the entire KG, KGNet significantly reduced training time and memory usage while maintaining comparable or improved accuracy. The KGNet source-code1is available for further study.

Knowledge from diverse application domains is organized as knowledge graphs (KGs) that are stored in RDF engines accessible in the web via SPARQL endpoints. Expressing a well-formed SPARQL query requires information about the graph structure and the exact URIs of its components, which is impractical for the average user. Question answering (QA) systems assist by translating natural language questions to SPARQL. Existing QA systems are typically based on application-specific human-curated rules, or require prior information, expensive pre-processing and model adaptation for each targeted KG. Therefore, they are hard to generalize to a broad set of applications and KGs. In this paper, we propose KGQAn, a universal QA system that does not need to be tailored to each target KG. Instead of curated rules, KGQAn introduces a novel formalization of question understanding as a text generation problem to convert a question into an intermediate abstract representation via a neural sequence-to-sequence model. We also develop a just-in-time linker that maps at query time the abstract representation to a SPARQL query for a specific KG, using only the publicly accessible APIs and the existing indices of the RDF store, without requiring any pre-processing. Our experiments with several real KGs demonstrate that KGQAn is easily deployed and outperforms by a large margin the state-of-the-art in terms of quality of answers and processing time, especially for arbitrary KGs, unseen during the training.

This demo paper presents KGNet, a graph machine learning-enabled RDF engine. KGNet integrates graph machine learning (GML) models with existing RDF engines as query operators to support node classification and link prediction tasks. For easy integration, KGNet extends the SPARQL language with user-defined predicates to support the GML operators. We refer to this extension as SPARQL ML query. Our SPARQL ML query optimizer is in charge of optimizing the selection of the near-optimal GML models. The development of KGNet poses research opportunities in various areas spanning KG management. In the paper, we demonstrate the ease of integration between the RDF engines and GML models through the SPARQL ML inference query language. We present several real use cases of different GML tasks on real KGs. Using KGNet, users do not need to learn a new scripting language or have a deep understanding of GML methods. The audience will experience KGNet with different KGs and GML models, as shown in our demo video and Colab notebook.

AutoML systems build machine learning models automatically by performing a search over valid data transformations and learners, along with hyper-parameter optimization for each learner. Many AutoML systems use meta-learning to guide search for optimal pipelines. In this work, we present a novel meta-learning system called KGpip which (1) builds a database of datasets and corresponding pipelines by mining thousands of scripts with program analysis, (2) uses dataset embeddings to find similar datasets in the database based on its content instead of metadata-based features, (3) models AutoML pipeline creation as a graph generation problem, to succinctly characterize the diverse pipelines seen for a single dataset. KGpip's meta-learning is a sub-component for AutoML systems. We demonstrate this by integrating KGpip with two AutoML systems. Our comprehensive evaluation using 121 datasets, including those used by the state-of-the-art systems, shows that KGpip significantly outperforms these systems.

Data science growing success relies on knowing where a relevant dataset exists, understanding its impact on a specific task, finding ways to enrich a dataset, and leveraging insights derived from it. With the growth of open data initiatives, data scientists need an extensible set of effective discovery operations to find relevant data from their enterprise datasets accessible via data discovery systems or open datasets accessible via data portals. Existing portals and systems suffer from limited discovery support and do not track the use of a dataset and insights derived from it. We will demonstrate KGLac, a system that captures metadata and semantics of datasets to construct a knowledge graph (GLac) interconnecting data items, e.g., tables and columns. KGLac supports various data discovery operations via SPARQL queries for table discovery, unionable and joinable tables, plus annotation with related derived insights. We harness a broad range of Machine Learning (ML) approaches with GLac to enable automatic graph learning for advanced and semantic data discovery. The demo will showcase how KGLac facilitates data discovery and enrichment while developing an ML pipeline to evaluate potential gender salary bias in IT jobs.

Employees that spend more time finding relevant data than analyzing it suffer from a data discovery problem. The large volume of data in enterprises, and sometimes the lack of knowledge of the schemas aggravates this problem. Similar to how we navigate the Web, we propose to identify semantic links that assist analysts in their discovery tasks. These links relate tables to each other, to facilitate navigating the schemas. They also relate data to external data sources, such as ontologies and dictionaries, to help explain the schema meaning. We materialize the links in an enterprise knowledge graph, where they become available to analysts. The main challenge is how to find pairs of objects that are semantically related. We propose SEMPROP, a DAG of different components that find links based on syntactic and semantic similarities. SEMPROP is commanded by a semantic matcher which leverages word embeddings to find objects that are semantically related. We introduce coherent group, a technique to combine word embeddings that works better than other state of the art combination alternatives. We implement SEMPROP as part of Aurum, a data discovery system we are building, and conduct user studies, real deployments and a quantitative evaluation to understand the benefits of links for data discovery tasks, as well as the benefits of SEMPROP and coherent groups to find those links.

In order for an enterprise to gain insight into its internal business and the changing outside environment, it is essential to provide the relevant data for in-depth analysis. Enterprise data is usually scattered across departments and geographic regions and is often inconsistent. Data scientists spend the majority of their time finding, preparing, integrating, and cleaning relevant data sets. Data Civilizer is an end-to-end data preparation system. In this paper, we present the complete system, focusing on our new workflow engine, a superior system for entity matching and consolidation, and new cleaning tools. Our workflow engine allows data scientists to author, execute and retrofit data preparation pipelines of different data discovery and cleaning services. Our end-to-end demo scenario is based on data from the MIT data warehouse and e-commerce data sets.

Solving business problems increasingly requires going beyond the limits of a single data processing platform (platform for short), such as Hadoop or a DBMS. As a result, organizations typically perform tedious and costly tasks to juggle their code and data across different platforms. Addressing this pain and achieving automatic cross-platform data processing is quite challenging: finding the most efficient platform for a given task requires quite good expertise for all the available platforms. We present R heem , a general-purpose cross-platform data processing system that decouples applications from the underlying platforms. It not only determines the best platform to run an incoming task, but also splits the task into subtasks and assigns each subtask to a specific platform to minimize the overall cost (e.g., runtime or monetary cost). It features (i) an interface to easily compose data analytic tasks; (ii) a novel cost-based optimizer able to find the most efficient platform in almost all cases; and (iii) an executor to efficiently orchestrate tasks over different platforms. As a result, it allows users to focus on the business logic of their applications rather than on the mechanics of how to compose and execute them. Using different real-world applications with R heem , we demonstrate how cross-platform data processing can accelerate performance by more than one order of magnitude compared to single-platform data processing.

Applications in life sciences, decentralized social networks, Internet of Things, and statistical linked dataspaces integrate data from multiple decentralized RDF graphs via SPARQL queries. Several approaches have been proposed to optimize query processing over a small number of heterogeneous data sources by utilizing schema information. In the case of schema similarity and interlinks among sources, these approaches cause unnecessary data retrieval and communication, leading to poor scalability and response time. This paper addresses these limitations and presents Lusail, a system for scalable and efficient SPARQL query processing over decentralized graphs. Lusail achieves scalability and low query response time through various optimizations at compile and run times. At compile time, we use a novel locality-aware query decomposition technique that maximizes the number of query triple patterns sent together to a source based on the actual location of the instances satisfying these triple patterns. At run time, we use selectivity-awareness and parallel query execution to reduce network latency and to increase parallelism by delaying the execution of subqueries expected to return large results. We evaluate Lusail using real and synthetic benchmarks, with data sizes up to billions of triples on an in-house cluster and a public cloud. We show that Lusail outperforms state-of-the-art systems by orders of magnitude in terms of scalability and response time.

Finding relevant data for a specific task from the numerous data sources available in any organization is a daunting task. This is not only because of the number of possible data sources where the data of interest resides, but also due to the data being scattered all over the enterprise and being typically dirty and inconsistent. In practice, data scientists are routinely reporting that the majority (more than 80%) of their effort is spent finding, cleaning, integrating, and accessing data of interest to a task at hand. We propose to demonstrate DATA CIVILIZER to ease the pain faced in analyzing data "in the wild". DATA CIVILIZER is an end-to-end big data management system with components for data discovery, data integration and stitching, data cleaning, and querying data from a large variety of storage engines, running in large enterprises.

There has been a proliferation of datasets available as interlinked RDF data accessible through SPARQL endpoints. This has led to the emergence of various applications in life science, distributed social networks, and Internet of Things that need to integrate data from multiple endpoints.

The RDF data model allows publishing interlinked RDF datasets, where each dataset is independently maintained and is queryable via a SPARQL endpoint. Many applications would benefit from querying the resulting large, decentralized, geo-distributed graph through a federated SPARQL query processor. A crucial factor for good performance in federated query processing is pushing as much computation as possible to the local endpoints. Surprisingly, existing federated SPARQL engines are not effective at this task since they rely only on schema information. Consequently, they cause unnecessary data retrieval and communication, leading to poor scalability and response time. This paper addresses these limitations and presents Lusail , a scalable and efficient federated SPARQL system for querying large RDF graphs that are geo-distributed on different endpoints. Lusail uses a novel query rewriting algorithm to push computation to the local endpoints by relying on information about the RDF instances and not only the schema. The query rewriting algorithm has the additional advantage of exposing parallelism in query processing, which Lusail exploits through advanced scheduling at query run time. Our experiments on billions of triples of real and synthetic data show that Lusail outperforms state-of-the-art systems by orders of magnitude in terms of scalability and response time.

This paper presents APlug, a framework for automatic tuning of large scale applications of many independent tasks. APlug suggests the best decomposition of the original computation into smaller tasks and the best number of CPUs to use, in order to meet user-specific constraints. We show that the problem is not trivial because there is large variability in the execution time of tasks, and it is possible for a task to occupy a CPU by performing useless computations. APlug collects a sample of task execution times and builds a model, which is then used by a discrete event simulator to calculate the optimal parameters. We provide a C++ API and a stand-alone implementation of APlug, and we integrate it with three typical applications from computational chemistry, bioinformatics, and data mining. A scenario for optimizing resources utilization is used to demonstrate our framework. We run experiments on 16,384 CPUs on a supercomputer, 480 cores on a Linux cluster and 80 cores on Amazon EC2, and show that APlug is very accurate with minimal overhead.

Strings and applications using them are proliferating in science and business. Currently, strings are stored in file systems and processed using ad-hoc procedural code. Existing techniques are not flexible and cannot efficiently handle complex queries or large datasets. In this paper, we demonstrate StarDB, a distributed database system for analytics on strings. StarDB hides data and system complexities and allows users to focus on analytics. It uses a comprehensive set of parallel string operations and provides a declarative query language to solve complex queries. StarDB automatically tunes itself and runs with over 90% efficiency on supercomputers, public clouds, clusters, and workstations. We test StarDB using real datasets that are 2 orders of magnitude larger than the datasets reported by previous works.

Providing the ability to elastically use more or fewer servers on demand (scale out and scale in) as the load varies is essential for database management systems (DBMSes) deployed on today's distributed computing platforms, such as the cloud. This requires solving the problem of dynamic (online) data placement, which has so far been addressed only for workloads where all transactions are local to one sever. In DBMSes where ACID transactions can access more than one partition, distributed transactions represent a major performance bottleneck. Scaling out and spreading data across a larger number of servers does not necessarily result in a linear increase in the overall system throughput, because transactions that used to access only one server may become distributed. In this paper we present Accordion, a dynamic data placement system for partition-based DBMSes that support ACID transactions (local or distributed). It does so by explicitly considering the affinity between partitions, which indicates the frequency in which they are accessed together by the same transactions. Accordion estimates the capacity of a server by explicitly considering the impact of distributed transactions and affinity on the maximum throughput of the server. It then integrates this estimation in a mixed-integer linear program to explore the space of possible configurations and decide whether to scale out. We implemented Accordion and evaluated it using H-Store, a shared-nothing in-memory DBMS. Our results using the TPC-C and YCSB benchmarks show that Accordion achieves benefits compared to alternative heuristics of up to an order of magnitude reduction in the number of servers used and in the amount of data migrated.

On-line transaction processing (OLTP) database management systems (DBMSs) often serve time-varying workloads due to daily, weekly or seasonal fluctuations in demand, or because of rapid growth in demand due to a company's business success. In addition, many OLTP workloads are heavily skewed to "hot" tuples or ranges of tuples. For example, the majority of NYSE volume involves only 40 stocks. To deal with such fluctuations, an OLTP DBMS needs to be elastic; that is, it must be able to expand and contract resources in response to load fluctuations and dynamically balance load as hot tuples vary over time. This paper presents E-Store, an elastic partitioning framework for distributed OLTP DBMSs. It automatically scales resources in response to demand spikes, periodic events, and gradual changes in an application's workload. E-Store addresses localized bottlenecks through a two-tier data placement strategy: cold data is distributed in large chunks, while smaller ranges of hot tuples are assigned explicitly to individual nodes. This is in contrast to traditional single-tier hash and range partitioning strategies. Our experimental evaluation of E-Store shows the viability of our approach and its efficacy under variations in load across a cluster of machines. Compared to single-tier approaches, E-Store improves throughput by up to 130% while reducing latency by 80%.

Motifs are frequent patterns used to identify biological functionality in genomic sequences, periodicity in time series, or user trends in web logs. In contrast to a lot of existing work that focuses on collections of many short sequences, modern applications require mining of motifs in one very long sequence (i.e., in the order of several gigabytes). For this case, there exist statistical approaches that are fast but inaccurate; or combinatorial methods that are sound and complete. Unfortunately, existing combinatorial methods are serial and very slow. Consequently, they are limited to very short sequences (i.e., a few megabytes), small alphabets (typically 4 symbols for DNA sequences), and restricted types of motifs.

A wide range of applications, including bioinformatics, time series, and log analysis, depend on the identification of repetitions in very long sequences. The problem of finding maximal pairs subsumes most important types of repetition-finding tasks. Existing solutions require both the input sequence and its index (typically an order of magnitude larger than the input) to fit in memory. Moreover, they are serial algorithms with long execution time. Therefore, they are limited to small datasets, despite the fact that modern applications demand orders of magnitude longer sequences. In this paper we present RACE, a parallel system for finding maximal pairs in very long sequences. RACE supports parallel execution on stand-alone multicore systems, in addition to scaling to thousands of nodes on clusters or supercomputers. RACE does not require the input or the index to fit in memory; therefore, it supports very long sequences with limited memory. Moreover, it uses a novel array representation that allows for cache-efficient implementation. RACE is particularly suitable for the cloud (e.g., Amazon EC2) because, based on availability, it can scale elastically to more or fewer machines during its execution. Since scaling out introduces overheads, mainly due to load imbalance, we propose a cost model to estimate the expected speedup, based on statistics gathered through sampling. The model allows the user to select the appropriate combination of cloud resources based on the provider's prices and the required deadline. We conducted extensive experimental evaluation with large real datasets and large computing infrastructures. In contrast to existing methods, RACE can handle the entire human genome on a typical desktop computer with 16GB RAM. Moreover, for a problem that takes 10 hours of serial execution, RACE finishes in 28 seconds using 2,048 nodes on an IBM BlueGene/P supercomputer.

The suffix tree is a data structure for indexing strings. It is used in a variety of applications such as bioinformatics, time series analysis, clustering, text editing and data compression. However, when the string and the resulting suffix tree are too large to fit into the main memory, most existing construction algorithms become very inefficient. This paper presents a disk-based suffix tree construction method, called Elastic Range (ERa), which works efficiently with very long strings that are much larger than the available memory. ERa partitions the tree construction process horizontally and vertically and minimizes I/Os by dynamically adjusting the horizontal partitions independently for each vertical partition, based on the evolving shape of the tree and the available memory. Where appropriate, ERa also groups vertical partitions together to amortize the I/O cost. We developed a serial version; a parallel version for shared-memory and shared-disk multi-core systems; and a parallel version for shared-nothing architectures. ERa indexes the entire human genome in 19 minutes on an ordinary desktop computer. For comparison, the fastest existing method needs 15 minutes using 1024 CPUs on an IBM BlueGene supercomputer.

Client/server information systems use caching techniques to reduce the volume of transmitted data as well as response time and, especially in the case of systems with mobile clients, to reduce energy consumptions. Updating the server database might cause inconsistencies between server data and cached data. Guaranteeing consistency at least demands to invalidate outdated caches. To avoid invalidation of caches that are not affected by a particular update one must check the relevancy of each update for each cache. It has been proven, that this can only be done on a stateful server.

Most mobile and embedded devices are battery powered. Hence, their uptime depends on the energy consumption of the used components. Developers made severe effort to optimize hardware components in order to reduce their energy consumption. However, in this paper we show that one also has to consider energy awareness in terms of software. In this study we focus on sorting algorithms, which are not only used directly by the user of a device but also very often implicitly by other algorithms. Our experiments show,that different sorting algorithms consume different amounts of energy. In detail, the experiments show that there is no direct correlation between the time complexity of an algorithm and its energy consumption.

Nowadays database systems (DBSs) are state-of-the art for managing complex data in information systems. Unfortunately, DBSs are not aware of the context of their usage. Query results are retrieved without considering the context of the user/device issuing the query. In many cases, this feature is added by a context managing middle-ware. This paper outlines a scenario for context aware mobile services. In the paper we introduce an XML-based XREAL model for formalizing contextual information and queries highlighted in the scenario and discuss how XREAL can be implemented in a modern database system. By utilizing XREAL, it is possible to realize context aware mobile information systems based on the available DBSs without additional middle-wares.