The emergence of novel hardware accelerators has powered the tremendous growth of machine learning in recent years. These accelerators deliver incomparable performance gains in processing high-volume ...matrix operators, particularly matrix multiplication, a core component of neural network training and inference. In this work, we explored opportunities of accelerating database systems using NVIDIA's Tensor Core Units (TCUs). We present TCUDB, a TCU-accelerated query engine processing a set of query operators including natural joins and group-by aggregates as matrix operators within TCUs. Matrix multiplication was considered inefficient in the past; however, this strategy has remained largely unexplored in conventional GPU-based databases, which primarily rely on vector or scalar processing. We demonstrate the significant performance gain of TCUDB in a range of real-world applications including entity matching, graph query processing, and matrix-based data analytics. TCUDB achieves up to 288x speedup compared to a baseline GPU-based query engine.
As the primary approach to deriving decision-support insights, automated recurring routine analytic jobs account for a major part of cluster resource usages in modern enterprise data warehouses. ...These recurring routine jobs usually have stringent schedule and deadline determined by external business logic, and thus cause dreadful resource skew and severe resource over-provision in the cluster. In this paper, we present Grosbeak, a novel data warehouse that supports resource-aware incremental computing to process recurring routine jobs, smooths the resource skew, and optimizes the resource usage. Unlike batch processing in traditional data warehouses, Grosbeak leverages the fact that data is continuously ingested. It breaks an analysis job into small batches that incrementally process the progressively available data, and schedules these small-batch jobs intelligently when the cluster has free resources. In this demonstration, we showcase Grosbeak using real-world analysis pipelines. Users can interact with the data warehouse by registering recurring queries and observing the incremental scheduling behavior and smoothed resource usage pattern.
Abstract
The Drug-Gene Interaction Database (DGIdb, www.dgidb.org) is a web resource that provides information on drug-gene interactions and druggable genes from publications, databases, and other ...web-based sources. Drug, gene, and interaction data are normalized and merged into conceptual groups. The information contained in this resource is available to users through a straightforward search interface, an application programming interface (API), and TSV data downloads. DGIdb 4.0 is the latest major version release of this database. A primary focus of this update was integration with crowdsourced efforts, leveraging the Drug Target Commons for community-contributed interaction data, Wikidata to facilitate term normalization, and export to NDEx for drug-gene interaction network representations. Seven new sources have been added since the last major version release, bringing the total number of sources included to 41. Of the previously aggregated sources, 15 have been updated. DGIdb 4.0 also includes improvements to the process of drug normalization and grouping of imported sources. Other notable updates include the introduction of a more sophisticated Query Score for interaction search results, an updated Interaction Score, the inclusion of interaction directionality, and several additional improvements to search features, data releases, licensing documentation and the application framework.
We introduce F-IVM, a unified incremental view maintenance (IVM) approach for a variety of tasks, including gradient computation for learning linear regression models over joins, matrix chain ...multiplication, and factorized evaluation of conjunctive queries.
F-IVM is a higher-order IVM algorithm that reduces the maintenance of the given task to the maintenance of a hierarchy of increasingly simpler views. The views are functions mapping keys, which are tuples of input data values, to payloads, which are elements from a task-specific ring. Whereas the computation over the keys is the same for all tasks, the computation over the payloads depends on the task. F-IVM achieves efficiency by factorizing the computation of the keys, payloads, and updates.
We implemented F-IVM as an extension of DBToaster. We show in a range of scenarios that it can outperform classical first-order IVM, DBToaster's fully recursive higher-order IVM, and plain recomputation by orders of magnitude while using less memory.
Summary
Variation and tradeoffs within and among plant traits are increasingly being harnessed by empiricists and modelers to understand and predict ecosystem processes under changing environmental ...conditions. While fine roots play an important role in ecosystem functioning, fine‐root traits are underrepresented in global trait databases. This has hindered efforts to analyze fine‐root trait variation and link it with plant function and environmental conditions at a global scale. This Viewpoint addresses the need for a centralized fine‐root trait database, and introduces the Fine‐Root Ecology Database (FRED, http://roots.ornl.gov) which so far includes > 70 000 observations encompassing a broad range of root traits and also includes associated environmental data. FRED represents a critical step toward improving our understanding of below‐ground plant ecology. For example, FRED facilitates the quantification of variation in fine‐root traits across root orders, species, biomes, and environmental gradients while also providing a platform for assessments of covariation among root, leaf, and wood traits, the role of fine roots in ecosystem functioning, and the representation of fine roots in terrestrial biosphere models. Continued input of observations into FRED to fill gaps in trait coverage will improve our understanding of changes in fine‐root traits across space and time.
BullFrog is a relational DBMS that supports single-step schema migrations --- even those that are backwards incompatible --- without downtime, and without need for advanced warning. When a schema ...migration is submitted, BullFrog initiates a logical switch to the new schema, but physically migrates affected data lazily, as it is accessed by incoming transactions. BullFrog's internal concurrency control algorithms and data structures enable concurrent processing of schema migration operations with post-migration transactions, while ensuring exactly-once migration of all old data into the physical layout required by the new schema. BullFrog is implemented as an open source extension to PostgreSQL. Experiments using this prototype over a TPC-C based workload (supplemented to include schema migrations) show that BullFrog can achieve zero-downtime migration to non-trivial new schemas with near-invisible impact on transaction throughput and latency.
Abstract
RNAcentral is a comprehensive database of non-coding RNA (ncRNA) sequences, collating information on ncRNA sequences of all types from a broad range of organisms. We have recently added a ...new genome mapping pipeline that identifies genomic locations for ncRNA sequences in 296 species. We have also added several new types of functional annotations, such as tRNA secondary structures, Gene Ontology annotations, and miRNA-target interactions. A new quality control mechanism based on Rfam family assignments identifies potential contamination, incomplete sequences, and more. The RNAcentral database has become a vital component of many workflows in the RNA community, serving as both the primary source of sequence data for academic and commercial groups, as well as a source of stable accessions for the annotation of genomic and functional features. These examples are facilitated by an improved RNAcentral web interface, which features an updated genome browser, a new sequence feature viewer, and improved text search functionality. RNAcentral is freely available at https://rnacentral.org.
Configuration tuning is vital to optimize the performance of database management system (DBMS). It becomes more tedious and urgent for cloud databases (CDB) due to the diverse database instances and ...query workloads, which make the database administrator (DBA) incompetent. Although there are some studies on automatic DBMS configuration tuning, they have several limitations. Firstly, they adopt a pipelined learning model but cannot optimize the overall performance in an end-to-end manner. Secondly, they rely on large-scale high-quality training samples which are hard to obtain. Thirdly, there are a large number of knobs that are in continuous space and have unseen dependencies, and they cannot recommend reasonable configurations in such high-dimensional continuous space. Lastly, in cloud environment, they can hardly cope with the changes of hardware configurations and workloads, and have poor adaptability. To address these challenges, we design an end-to-end automatic CDB tuning system, CDBTune, using deep reinforcement learning (RL). CDBTune utilizes the deep deterministic policy gradient method to find the optimal configurations in high-dimensional continuous space. CDBTune adopts a try-and-error strategy to learn knob settings with a limited number of samples to accomplish the initial training, which alleviates the difficulty of collecting massive high-quality samples. CDBTune adopts the reward-feedback mechanism in RL instead of traditional regression, which enables end-to-end learning and accelerates the convergence speed of our model and improves efficiency of online tuning. We conducted extensive experiments under 6 different workloads on real cloud databases to demonstrate the superiority of CDBTune. Experimental results showed that CDBTune had a good adaptability and significantly outperformed the state-of-the-art tuning tools and DBA experts.
This open access book aims to educate data space designers to understand what is required to create a successful data space. It explores cutting-edge theory, technologies, methodologies, and best ...practices for data spaces for both industrial and personal data and provides the reader with a basis for understanding the design, deployment, and future directions of data spaces. The book captures the early lessons and experience in creating data spaces. It arranges these contributions into three parts covering design, deployment, and future directions respectively. The first part explores the design space of data spaces. The single chapters detail the organisational design for data spaces, data platforms, data governance federated learning, personal data sharing, data marketplaces, and hybrid artificial intelligence for data spaces. The second part describes the use of data spaces within real-world deployments. Its chapters are co-authored with industry experts and include case studies of data spaces in sectors including industry 4.0, food safety, FinTech, health care, and energy. The third and final part details future directions for data spaces, including challenges and opportunities for common European data spaces and privacy-preserving techniques for trustworthy data sharing. The book is of interest to two primary audiences: first, researchers interested in data management and data sharing, and second, practitioners and industry experts engaged in data-driven systems where the sharing and exchange of data within an ecosystem are critical.
PDF
