Access To Metadata Research Articles

X Meta : SSD-HDD-hybrid Optimization for Metadata Maintenance of Cloud-scale Object Storage

Object storage has been widely used in the cloud. Traditionally, the size of object metadata is much smaller than that of object data, and thus existing object storage systems (such as Ceph and Oasis) can place object data and metadata, respectively, on hard disk drives (HDDs) and solid-state drives (SSDs) to achieve high I/O performance at a low monetary cost. Currently, however, a wide range of cloud applications organize their data as large numbers of small objects of which the data size is close to (or even smaller than) the metadata size, thus greatly increasing the cost if placing all metadata on expensive SSDs. This article presents x Meta , an SSD-HDD-hybrid optimization for metadata maintenance of cloud-scale object storage. We observed that a substantial portion of the metadata of small objects is rarely accessed and thus can be stored on HDDs with little performance penalty. Therefore, x Meta first classifies the hot and cold metadata based on the frequency of metadata accesses of upper-layer applications and then adaptively stores the hot metadata on SSDs and the cold metadata on HDDs. We also propose a merging mechanism for hot metadata to further improve the efficiency of SSD storage and optimize range key query and insertion for hot metadata by designing composite keys. We have integrated the x Meta metadata service with Ceph to realize a high-performance, low-cost object store (called xCeph). The extensive evaluation shows that xCeph outperforms the original Ceph by an order of magnitude in the space requirement of SSD storage, while improving the throughput by up to 2.7×.

Hybrid deduplication system with content-based cache for cloud environment

Primary storage deduplication systems are performance sensitive. Their performance depends upon two factors — metadata access for duplicate detection and strategy for elimination of duplicate data. Various approaches for duplicate detection through suitable caching mechanisms have been proposed in the literature. Most of the approaches assumed that the primary workloads exhibit strong temporal locality. Whereas, this cannot be assumed in the context of Cloud as the workloads locality does not exist with interferences among different workloads on the same system. Duplicate content among the data blocks with different addresses lead to an inefficient utilization of the data cache. In this context, applying deduplication causes sharing of the data blocks among the clients with different access patterns and frequencies. In this situation, LRU cache, which considers only the recency of the references, is not appropriate. In this paper, Hybrid Deduplication System (HDS) containing the content-based cache with a new replacement policy — Modified Adaptive Replacement Cache (ARC), is proposed. The proposed system is simulated in the Linux environment using three different types of FIU traces. Effectiveness of the system is compared with a full deduplication system. Experimental results show that the system has performed consistently better than the full deduplication system in reducing the metadata overhead for all of the three data sets.

Abstract 6910: CancerModels.org: An open global cancer research platform for patient-derived cancer models

Abstract CancerModels.Org (www.cancermodels.org) is a research platform that standardises, harmonizes and integrates the complex and diverse data associated with Patient-Derived Cancer Models (PDCMs) for the cancer community. The portal publishes over 7500 models - covering patient-derived xenografts (PDX), organoids and cell lines - across 13 cancer types, including rare pediatric PDX models and models from minority ethnic backgrounds. A total of over 90 million data points are made available across a variety of data types, such as clinical metadata, molecular data and treatment-based information, which makes CancerModels.Org the largest free-to-consumer and open-access resource of this kind. Over the course of the last twelve months, the platform has been enhanced with new functionality and an updated user interface to cater for a more varied set of use cases. Users can now search for models of interest by exploring molecular data summaries for models of specific cancer types, as well as by using the intuitive search and faceted filtering options of the web interface. The data is also accessible via REST API, hence enabling offline analyses. The underpinning data model was augmented with additional dimensions and covers gene expression, gene mutation, copy number alteration, biomarkers, patient treatment and drug dosing studies. For an improved prioritization of PDCMs we performed knowledge enrichment by linking to external resources, such as publication platforms, cancer-specific annotation tools (COSMIC, CIViC, OncoMX, OpenCRAVAT, ClinGen), and raw data archives (ENA, EGA, GEO, dbGAP). More recently, we added immune-related features of tumors (HLA, MSI, TMB) and image data (also accessible via EMBL-EBI’s BioImage archive). Finally, to streamline model and data submission, we made available a Metadata dictionary and a Metadata validation service. Future work will focus on the curation of rare models with rich accessible metadata and data, new data visualizations, integration of new data types and resources, as well as devising a model quality rating using user feedback. These developments will maximize utility and improve reusability of models and data, and reduce barriers to model and data sharing. CancerModels.Org is deeply integrated into the general patient-derived cancer models landscape, underpinning or contributing to the data and/or software infrastructure of some of the long-running consortia, such as EUROPDX and PDXNet. The resource is freely available under an Apache 2.0 license (https://github.com/PDCMFinder). Citation Format: Zinaida Perova, Mauricio Martinez, Tushar Mandloi, Marcelo Rios Almanza, Steven Neuhauser, Dale Degley, Debra Krupke, Carol Bult, Helen Parkinson. CancerModels.org: An open global cancer research platform for patient-derived cancer models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6910.

RHPM: Using Relative Hotness to Guide Page Migration for Hybrid Memory Systems

Modern computing systems and data-intensive applications are eager for larger and faster memory. Building hybrid memory systems with different memory technologies has become a dominant trend to satisfy these demands. For hybrid memory systems, page migration schemes that dynamically migrate frequently accessed hot pages into faster memory are crucial for improving performance. However, existing migration schemes are either too aggressive, resulting in unnecessary extra traffic, or too conservative to quickly adapt to changes in access patterns. Besides, the extra latency introduced by querying metadata is often ignored or handled in an unscalable manner. In this paper, we propose a Relative Hotness Page Migration (RHPM) strategy, which discovers hot pages in a set of pages by competing with each other rather than comparing with a threshold. The migration is performed only when a new page wins the competition. To overlap latency due to access metadata, RHPM fetches metadata and data in parallel. In addition, it enables a small metadata buffer to speed up metadata access. Compared to the state-of-the-art scheme, RHPM requires only 1/512 of the on-chip capacity, significantly reducing on-chip hardware overhead. Evaluation of RHPM with simulations of 25 workloads shows that RHPM outperforms the state-of-the-art scheme by an average of 13.34% in performance and saves 44.19% on energy, demonstrating better resilience to changes in access patterns.

FAIR in action: Brain-CODE - A neuroscience data sharing platform to accelerate brain research.

The effective sharing of health research data within the healthcare ecosystem can have tremendous impact on the advancement of disease understanding, prevention, treatment, and monitoring. By combining and reusing health research data, increasingly rich insights can be made about patients and populations that feed back into the health system resulting in more effective best practices and better patient outcomes. To achieve the promise of a learning health system, data needs to meet the FAIR principles of findability, accessibility, interoperability, and reusability. Since the inception of the Brain-CODE platform and services in 2012, the Ontario Brain Institute (OBI) has pioneered data sharing activities aligned with FAIR principles in neuroscience. Here, we describe how Brain-CODE has operationalized data sharing according to the FAIR principles. Findable-Brain-CODE offers an interactive and itemized approach for requesters to generate data cuts of interest that align with their research questions. Accessible-Brain-CODE offers multiple data access mechanisms. These mechanisms-that distinguish between metadata access, data access within a secure computing environment on Brain-CODE and data access via export will be discussed. Interoperable-Standardization happens at the data capture level and the data release stage to allow integration with similar data elements. Reusable - Brain-CODE implements several quality assurances measures and controls to maximize data value for reusability. We will highlight the successes and challenges of a FAIR-focused neuroinformatics platform that facilitates the widespread collection and sharing of neuroscience research data for learning health systems.

The unseen: revealing the blind production procedure and experience for NP data

A unique experiment was conducted by the STAR Collaboration in 2018 to investigate differences between collisions of nuclear isobars, a potential key to unraveling one of the physics mysteries in our field: why the universe is made predominantly of matter. Enhancing the credibility of findings was deemed to hinge on blinding analyzers from knowing which dataset they were examining, necessitating efforts by the data production team to investigate and implement new (in our field) blinding practices. With nearly two decades of established machinery intended to provide open data and metadata access in STAR, the breadth of details to consider for a successful blinding process was substantial.In this paper, we will review the experience of the first-ever species-blind data production in high energy collider physics. Considerations of what needed blinding, how to blind it, and from whom, will be discussed. Practical impositions on the selection of data to produce and how to produce it will also be highlighted as tell-tales can arise even where there were operational efforts to avoid them. With the appropriate blinding achieved, analyzers and reviewers have been empowered to focus on the physics of interest.

Data sharing in neurodegenerative disease research: challenges and learnings from the innovative medicines initiative public-private partnership model.

Efficient data sharing is hampered by an array of organizational, ethical, behavioral, and technical challenges, slowing research progress and reducing the utility of data generated by clinical research studies on neurodegenerative diseases. There is a particular need to address differences between public and private sector environments for research and data sharing, which have varying standards, expectations, motivations, and interests. The Neuronet data sharing Working Group was set up to understand the existing barriers to data sharing in public-private partnership projects, and to provide guidance to overcome these barriers, by convening data sharing experts from diverse projects in the IMI neurodegeneration portfolio. In this policy and practice review, we outline the challenges and learnings of the WG, providing the neurodegeneration community with examples of good practices and recommendations on how to overcome obstacles to data sharing. These obstacles span organizational issues linked to the unique structure of cross-sectoral, collaborative research initiatives, to technical issues that affect the storage, structure and annotations of individual datasets. We also identify sociotechnical hurdles, such as academic recognition and reward systems that disincentivise data sharing, and legal challenges linked to heightened perceptions of data privacy risk, compounded by a lack of clear guidance on GDPR compliance mechanisms for public-private research. Focusing on real-world, neuroimaging and digital biomarker data, we highlight particular challenges and learnings for data sharing, such as data management planning, development of ethical codes of conduct, and harmonization of protocols and curation processes. Cross-cutting solutions and enablers include the principles of transparency, standardization and co-design - from open, accessible metadata catalogs that enhance findability of data, to measures that increase visibility and trust in data reuse.

SMURF: Efficient and Scalable Metadata Access for Distributed Applications

In parallel with big data processing and analysis dominating the usage of distributed and Cloud infrastructures, the demand for distributed metadata access and transfer has increased. The volume of data generated by many application domains exceeds petabytes, while the corresponding metadata amounts to terabytes or even more. This article proposes a novel solution for efficient and scalable metadata access for distributed applications across wide-area networks, dubbed SMURF. Our solution combines novel pipelining and concurrent transfer mechanisms with reliability, provides distributed continuum caching and semantic locality-aware prefetching strategies to sidestep fetching latency, and achieves scalable and high-performance metadata fetch/prefetch services in the Cloud. We incorporate the phenomenon of semantic locality awareness for increased prefetch prediction rate using real-life application I/O traces from Yahoo! Hadoop audit logs and propose a novel prefetch predictor. By effectively caching and prefetching metadata based on the access patterns, our continuum caching and prefetching mechanism significantly improves the local cache hit rate and reduces the average fetching latency. We replay approximately 20 Million metadata access operations from real audit traces, where SMURF achieves 90% accuracy during prefetch prediction and reduced the average fetch latency by 50% compared to the state-of-the-art mechanisms.

Law Reviews, Open Metadata and RSS Feeds

AbstractMost scientific publications have their metadata available as freely accessible and machine-readable information at CrossRef. However, student-edited law reviews have not followed suit with this practice. Consequently, a large part of legal research remains in a blind spot for scientometric analyses and tools. The present paper, by Andreas Nishikawa-Pacher, investigates whether the law reviews’ RSS feeds could serve as equivalent sources for open scholarly metadata. The suitability of RSS feeds from 51 student-edited law reviews (as indexed in Web of Science's Social Science Citation Index) was assessed with regards to three fictitious meta-scientific applications – namely (1) a ‘latest paper’-tool that lists the law reviews’ newest publications with links, abstracts and dates, (2) an author database and (3) a calculation of the mutual citation counts among different law reviews. This paper finds that only 21 law reviews offer functional RSS feeds, and while they were suitable for a basic ‘latest papers’-tool, due to their low metadata quality they cannot aid in generating an author database or in counting the mutual citations among law reviews. The result suggests that law reviews would be advised to adopt digital object identifiers (DOIs) and start depositing openly accessible metadata, for otherwise their scholarly impact and visibility will further decline.

CATCH: Cloud Data Acquisition through Comprehensive and Hybrid Approaches

With the development of Internet technology, cloud-based services have improved the availability and usability of resources. Among them, cloud storage services enable users to remotely store, access, or share data over a network. Therefore, digital forensic investigators need to collect data stored in remote servers to comprehensively understand a suspect's activities. Although several well-known commercial digital forensic tools provide features for cloud data acquisition in order to support this requirement, fewer studies have addressed whether they have full access to cloud resources and collect all the data as expected. In this regard, our findings from this work show that those commercial tools do not completely identify and collect data that are obviously available through dedicated clients (e.g., web-browsers and desktop/mobile apps). In this paper, we propose an investigative framework, CATCH (Cloud Data Acquisition through Comprehensive and Hybrid Approaches), which is composed of four steps (Authentication, Exploration, Filtering, and Collection). CATCH collects authentication data to access cloud resources and then, explores, filters, and collects all accessible metadata as well as contents from remote cloud servers by using Open and Internal APIs. To demonstrate our proposal, the CATCH framework is applied to collect a user's Microsoft OneDrive storage from digital forensics perspectives. We then evaluate data collection results generated from a self-developed tool based on the proposed framework, by comparing them to results from commercial digital forensic tools.

Metadata, Digital Museum Spaces and Accessibility for Persons with Impairments

This study examines the intersection of metadata and accessibility in the digital museum space. While virtual museums can remove physical barriers to heritage institutions, virtual spaces still have barriers for certain impairments. Work done in the field of digital libraries has shown that accessibility metadata can help to remove some of those barriers. This exploratory research project aims to provide understanding of the ways that metadata in virtual museum spaces is being used and how that metadata can improve accessibility in these spaces. How can universal design be best applied to metadata in digital museum catalogs and objects? What kinds of metadata in digital museum collections can potentially increase accessibility? Are museums currently including metadata that increases accessibility for people with disabilities and impairments? These questions were explored using a content-analysis approach applied to textual information collected from forty objects in four digital collections found on the publicly accessible webpages of Canadian institutions of cultural heritage.

DIESEL+: Accelerating Distributed Deep Learning Tasks on Image Datasets

We observe that data access and processing takes a significant amount of time in large-scale deep learning training tasks (DLTs) on image datasets. Three factors contribute to this problem: (1) the massive and recurrent accesses to large numbers of small files; (2) the repeated, expensive decoding computation on each image, and (3) the frequent communication between computation nodes and storage nodes. Existing work has addressed some aspects of these problems; however, no end-to-end solutions have been proposed. In this article, we propose DIESEL+, an all-in-one system which accelerates the entire I/O pipeline of deep learning training tasks. DIESEL+ contains several components: (1) local metadata snapshot; (2) per-task distributed caching; (3) chunk-wise shuffling; (4) GPU-assisted image decoding and (5) online region-of-interest (ROI) decoding. The metadata snapshot removes the bottleneck on metadata access in frequent reading of large numbers of files. The per-task distributed cache across the worker nodes of a DLT task to reduce the I/O pressure on the underlying storage. The chunk-based shuffle method converts small file reads into large chunk reads, so that the performance is improved without sacrificing the training accuracy. The GPU-assisted image decoding and the online ROI method minimize the image decoding workloads and reduce the cost of data movement between nodes. These techniques are seamlessly integrated into the system. In our experiments, DIESEL+ outperforms existing systems by a factor of two to three times on the overall training time.

IRO: Integrity–Reliability enhanced Ring ORAM

Memory security and reliability are two major design concerns in cloud computing systems. State-of-the-art memory security–reliability co-designs (e.g., Synergy) have achieved a good balance in performance, security, and reliability. However, these works merely rely on encryption to ensure data confidentiality, which cannot avoid information leakage from memory access patterns. Ring ORAM is an attractive confidentiality protection protocol to hide memory access patterns to the untrusted system. Unfortunately, it lacks memory integrity and reliability support, and it is incompatible with the security–reliability co-designs. A forced combination of Ring ORAM with the general integrity and reliability schemes would result in severe performance and storage overhead. In this paper, we propose IRO, an Integrity–Reliability enhanced Ring ORAM design. First, we analyze the inefficiency of general integrity protection methods and propose RIT, an integrity verification scheme for Ring ORAM that does not have extra integrity metadata access overhead. Then, we analyze the memory waste in Ring ORAM and propose Secure Replication, the reliability scheme that provides channel-level memory error resilience with very little storage overhead for replicas. We conduct cycle-accurate simulations to analyze the efficiency of our design. The results show that IRO increases 7.56% execution time of the baseline Ring ORAM in the case of limited MAC computation units. With enough computation resources, IRO can reduce the execution time of the baseline by 2.11%.

File Semantic Aware Primary Storage Deduplication System

The deployment of the primary storage deduplication system at the file level is a daunting challenge due to disk-bottleneck and data fragmentation problems. File semantics such as type and size can be used to reduce deduplication overhead. There is negligible data redundancy across different types of files compared to the same type of files. Hence, the application of the same deduplication method irrespective of file type causes wastage of computing resources. In this paper, the File Aware DeDuplicaton system is proposed. Files are partitioned based on the size as small and large files. Large files are categorized based on the data redundancy as high, low, and unpredictable file types. File type-specific deduplication approach is applied for each category of files. Separate index tables are maintained for each type and hash table for small files of all types. The FADD system is simulated in the Linux environment using two different types of FIU traces and some locally collected data sets. The effectiveness of FADD is compared with the full deduplication system and Hybrid Deduplication System based on the parameters – metadata access overhead, average segment length, and response time. The experimental results show that the system has performed consistently better for all input data sets.

Automatic Adaptation of Open Educational Resources: An Approach From a Multilevel Methodology Based on Students’ Preferences, Educational Special Needs, Artificial Intelligence and Accessibility Metadata

The need for adaptive e-learning environments that respond to learning variability is now a fundamental requirement in education, as it helps to ensure that students learn and pass their courses within a set time frame. Although guidelines, techniques and methods have been established in recent years to contribute to the development of accessible and adaptable e-learning environments that promote digital inclusion, their implementation is challenging due to the lack of knowledge of an adequate way to do it and because it is considered more of a technological competence for scholars in the area. In this context, automated support for adapting material that responds to the correct use of accessibility metadata not only provides a way to improve the description of adapted educational resources, but also facilitates their search according to the needs and preferences of students, particularly those with disabilities. In this article, we carry out a multilevel methodological proposal for the automatic adaptation of open educational resources, in order to provide a tool that contributes to the accessibility and correct use of their metadata in e-learning environments. A research is conducted with students with disabilities to establish their real needs and preferences, highlighting the need to strengthen the adequate description and coherent alternative text in images, the correct subtitling in videos and the conversion of audio to text, data that are relevant to our proposal. The research conducted aims to contribute with an automated support tool in the generation of accessible educational resources that are correctly labeled for search and reuse. This research also aims to support researchers in artificial intelligence applications to address challenges and opportunities in the field of virtual education, in addition to providing an overview that could help those who generate educational resources and maintain their interest in making them accessible.

The use of accessibility metadata in e-learning environments: a systematic literature review

E-learning environments constitute an essential element in education, as they help students to ensure their pass their courses and graduate on time. Although guidelines, techniques, and methods have been presented in some literature in recent years to contribute to the development of accessible e-learning environments that promote digital inclusion, their implementation is challenging. In this context, the use of accessibility metadata not only provides a way to enhance the description of adapted educational resources but also facilitates their search according to the needs and preferences of students, in particular those with disabilities. In this paper, a systematic review was conducted in order to provide the state of the art regarding the use of accessibility metadata in e-learning environments. A total of 746 documents were found during the period from 2012 to 2019, of which 31 were selected according to the inclusion and exclusion criteria relevant to our review. The findings revealed an intensive use of models and standards of accessibility in e-learning environments, however, using accessibility metadata remains underused. In fact, the evaluation of accessibility and adaptability through the use of its metadata was not found. The findings obtained also helped to establish challenges and opportunities in this research field as well as to provide an overview that could support those who generate educational resources to keep their interest in making them accessible.

SecNVM: An Efficient and Write-Friendly Metadata Crash Consistency Scheme for Secure NVM

Data security is an indispensable part of non-volatile memory (NVM) systems. However, implementing data security efficiently on NVM is challenging, since we have to guarantee the consistency of user data and the related security metadata. Existing consistency schemes ignore the recoverability of the SGX style integrity tree (SIT) and the access correlation between metadata blocks, thereby generating unnecessary NVM write traffic. In this article, we propose SecNVM, an efficient and write-friendly metadata crash consistency scheme for secure NVM. SecNVM utilizes the observation that for a lazily updated SIT, the lost tree nodes after a crash can be recovered by the corresponding child nodes in NVM. It reduces the SIT persistency overhead through a restrained write-back metadata cache and exploits the SIT inter-layer dependency for recovery. Next, leveraging the strong access correlation between the counter and DMAC, SecNVM improves the efficiency of security metadata access through a novel collaborative counter-DMAC scheme. In addition, it adopts a lightweight address tracker to reduce the cost of address tracking for fast recovery. Experiments show that compared to the state-of-the-art schemes, SecNVM improves the performance and decreases write traffic a lot, and achieves an acceptable recovery time.

Secure and Privacy-Respecting Documentation for Interactive Manufacturing and Quality Assurance

The automated documentation of work steps is a requirement of many modern manufacturing processes. Especially when it comes to important procedures such as safety critical screw connections or weld seams, the correct and complete execution of certain manufacturing steps needs to be properly supervised, e.g., by capturing video snippets of the worker to be checked in hindsight. Without proper technical and organizational safeguards, such documentation data carries the potential for covert performance monitoring to the disadvantage of employees. Naïve documentation architectures interfere with data protection requirements, and thus cannot expect acceptance of employees. In this paper we outline use cases for automated documentation and describe an exemplary system architecture of a workflow recognition and documentation system. We derive privacy protection goals that we address with a suitable security architecture based on hybrid encryption, secret-sharing among multiple parties and remote attestation of the system to prevent manipulation. We finally contribute an outlook towards problems and possible solutions with regards to information that can leak through accessible metadata and with regard to more modular system architectures, where more sophisticated remote attestation approaches are needed to ensure the integrity of distributed components.

Hadoop Perfect File: A fast and memory-efficient metadata access archive file to face small files problem in HDFS

HDFS faces several issues when it comes to handling a large number of small files. These issues are well addressed by archive systems, which combine small files into larger ones. They use index files to hold relevant information for retrieving a small file content from the big archive file. However, existing archive-based solutions require significant overheads when retrieving a file content since additional processing and I/Os are needed to acquire the retrieval information before accessing the actual file content, therefore, deteriorating the access efficiency. This paper presents a new archive file named Hadoop Perfect File (HPF). HPF minimizes access overheads by directly accessing metadata from the part of the index file containing the information. It consequently reduces the additional processing and I/Os needed and improves the access efficiency from archive files. Our index system uses two hash functions. Metadata records are distributed across index files using a dynamic hash function. We further build an order-preserving perfect hash function that memorizes the position of a small file's metadata record within the index file.

Pragmatic MDR: a metadata repository with bottom-up standardization of medical metadata through reuse

BackgroundThe variety of medical documentation often leads to incompatible data elements that impede data integration between institutions. A common approach to standardize and distribute metadata definitions are ISO/IEC 11179 norm-compliant metadata repositories with top-down standardization. To the best of our knowledge, however, it is not yet common practice to reuse the content of publicly accessible metadata repositories for creation of case report forms or routine documentation. We suggest an alternative concept called pragmatic metadata repository, which enables a community-driven bottom-up approach for agreeing on data collection models. A pragmatic metadata repository collects real-world documentation and considers frequent metadata definitions as high quality with potential for reuse.MethodsWe implemented a pragmatic metadata repository proof of concept application and filled it with medical forms from the Portal of Medical Data Models. We applied this prototype in two use cases to demonstrate its capabilities for reusing metadata: first, integration into a study editor for the suggestion of data elements and, second, metadata synchronization between two institutions. Moreover, we evaluated the emergence of bottom-up standards in the prototype and two medical data managers assessed their quality for 24 medical concepts.ResultsThe resulting prototype contained 466,569 unique metadata definitions. Integration into the study editor led to a reuse of 1836 items and item groups. During the metadata synchronization, semantic codes of 4608 data elements were transferred. Our evaluation revealed that for less complex medical concepts weak bottom-up standards could be established. However, more diverse disease-related concepts showed no convergence of data elements due to an enormous heterogeneity of metadata. The survey showed fair agreement (Kalpha = 0.50, 95% CI 0.43–0.56) for good item quality of bottom-up standards.ConclusionsWe demonstrated the feasibility of the pragmatic metadata repository concept for medical documentation. Applications of the prototype in two use cases suggest that it facilitates the reuse of data elements. Our evaluation showed that bottom-up standardization based on a large collection of real-world metadata can yield useful results. The proposed concept shall not replace existing top-down approaches, rather it complements them by showing what is commonly used in the community to guide other researchers.