Search

The cloud is the best method used for the utilization and organization of data. The cloud provides many resources for us via the internet. There are many technologies used in cloud computing systems; each one uses a different kind of protocols and methods. Many tasks can execute on different servers per second, which cannot execute on their computer. The most popular technologies used in the cloud system are Hadoop, Dryad, and another map reducing framework. Also, there are many tools used to optimize the performance of the cloud system, such as Cap3, HEP, and Cloudburst. This paper reviews in detail the cloud computing system, its used technologies, and the best technologies used with it according to multiple factors and criteria such as the procedure cost, speed cons and pros. Moreover, A comprehensive comparison of the tools used for the utilization of cloud computing systems is presented.

Time and budgetary resources are often a limiting factor in the collection of large-scale ecological data. If data collected by citizen scientists were comparable to data collected by researchers, it would allow for more efficient data collection over a broad geographic area. Here, we compare the quality of data on bat activity collected by citizens (high school students and teachers) and researchers. Both researchers and citizen scientists used the same comprehensive instructions when choosing study sites. We found no differences in total bat activity minutes recorded by citizens and researchers. Instead, citizen scientists collected data from a wider variety of habitats than researchers. Involvement of citizens also increased the geographical coverage of data collection, resulting in the northernmost documentation of the Nathusius’s pipistrelle so far in Finland. Therefore, bat research can benefit from the use of citizen science when participants are given precise instructions and calibrated data collection equipment. Citizen science projects also have other far-reaching benefits, increasing, for example, the scientific literacy and interest in natural sciences of citizens. Involving citizens in science projects also has the potential to enhance their willingness to conserve nature.Open Research StatementData are not yet provided, but will uploaded Dryad upon publication.

In October Jen Gibson started as the new Executive Director for the Dryad Data Repository. I used the opportunity to ask Jen a few questions about Dryad, challenges with data sharing, and ideas about moving Dryad forward. ...

Recent years have seen rapid growth in data-driven distributed systems, such as Hadoop MapReduce, Spark, and Dryad. However, the counterparts for high-performance or compute-intensive applications including large-scale optimizations, modeling, and simulations are still nascent. In this paper, we introduce DtCraft, a modern C++ based distributed execution engine to streamline the development of high-performance parallel applications. Users need no understanding of distributed computing and can focus on high-level developments, leaving difficult details, such as concurrency controls, workload distribution, and fault tolerance handled by our system transparently. We have evaluated DtCraft on both micro-benchmarks and large-scale optimization problems, and shown the promising performance from single multicore machines to clusters of computers. In a particular semiconductor design problem, we achieved ${30\times }$ speedup with 40 nodes and ${15\times }$ less development efforts over hand-crafted implementation.

There were mistakes in the data of three variables used in this study. The corrections listed below have minor effects on statistics and p-values, but do not change any conclusions. Correct data are stored in the Dryad repository ( https://doi.org/10.5061/dryad.4f8d2.2 ).

ORCID is a scientific infrastructure created to solve the problem of author name ambiguity. Over the years ORCID has also become a useful source for studying academic activities reported by researchers. Our objective in this research was to use ORCID to analyze one of these research activities: the publication of datasets. We illustrate how the identification of datasets that shared in researchers' ORCID profiles enables the study of the characteristics of the researchers who have produced them. To explore the relevance of ORCID to study data sharing practices we obtained all ORCID profiles reporting at least one dataset in their "works" list, together with information related to the individual researchers producing the datasets. The retrieved data was organized and analyzed in a SQL database hosted at CWTS. Our results indicate that DataCite is by far the most important data source for providing information about datasets recorded in ORCID. There is also a substantial overlap between DataCite records with other repositories (Figshare, Dryad, and Zenodo). The analysis of the distribution of researchers producing datasets shows that the top six countries with more data producers, also have a relatively higher percentage of people who have produced datasets out of total researchers with datasets than researchers in the total ORCID. By disciplines, researchers that belong to the areas of Natural Sciences and Medicine and Life Sciences are those with the largest amount of reported datasets. Finally, we observed that researchers who have started their PhD around 2015 published their first dataset earlier that those researchers that started their PhD before. The work concludes with some reflections of the possibilities of ORCID as a relevant source for research on data sharing practices., Paper accepted for publication in Scientometrics

Data are one of the primary outputs of science. Although certain subdisciplines of biology have pioneered efforts to ensure their long-term preservation and facilitate collaborations, data continue to disappear, owing mostly to technological, regulatory, and ideological hurdles. In this article, we describe the important steps toward proper data management and archiving and provide a critical discussion on the importance of long-term data conservation. We then illustrate the rise in data archiving through the Joint Data Archiving Policy and the Dryad Digital Repository. In particular, we discuss data integration and how the limited availability of large-scale data sets can hinder new discoveries. Finally, we propose solutions to increase the rate of data preservation, for example by generating mechanisms insuring proper data management and archiving, by providing training in data management, and by transforming the traditional role of research institutions and libraries as data generators toward managers and archivers.

This work was assisted through participation in “Next Generation Genetic Monitoring” Investigative Workshop at the National Institute for Mathematical and Biological Synthesis, sponsored by the National Science Foundation through NSF Award #DBI-1300426, with additional support from The University of Tennessee, Knoxville. Hawaiian fish community data were provided by the NOAA Pacific Islands Fisheries Science Center's Coral Reef Ecosystem Division (CRED) with funding from NOAA Coral Reef Conservation Program. O.E.G. was supported by the Marine Alliance for Science and Technology for Scotland (MASTS). A. C. and C. H. C. were supported by the Ministry of Science and Technology, Taiwan. P.P.-N. was supported by a Canada Research Chair in Spatial Modelling and Biodiversity. K.A.S. was supported by National Science Foundation (BioOCE Award Number 1260169) and the National Center for Ecological Analysis and Synthesis. All data used in this manuscript are available in DRYAD (https://doi.org/dx.doi.org/10.5061/dryad.qm288) and BCO-DMO (http://www.bco-dmo.org/project/552879). Biological diversity is a key concept in the life sciences and plays a fundamental role in many ecological and evolutionary processes. Although biodiversity is inherently a hierarchical concept covering different levels of organisation (genes, population, species, ecological communities and ecosystems), a diversity index that behaves consistently across these different levels has so far been lacking, hindering the development of truly integrative biodiversity studies. To fill this important knowledge gap we present a unifying framework for the measurement of biodiversity across hierarchical levels of organisation. Our weighted, information-based decomposition framework is based on a Hill number of order q = 1, which weights all elements in proportion to their frequency and leads to diversity measures based on Shannon’s entropy. We investigated the numerical behaviour of our approach with simulations and showed that it can accurately describe complex spatial hierarchical structures. To demonstrate the intuitive and straightforward interpretation of our diversity measures in terms of effective number of components (alleles, species, etc.) we applied the framework to a real dataset on coral reef biodiversity. We expect our framework will have multiple applications covering the fields of conservation biology, community genetics, and eco-evolutionary dynamics. Publisher PDF

Cloud is referred to as a collection of infrastructure services, such as Infrastructure as a service (IaaS) and Platform as a service (PaaS), which are made available to us for utilization by various organizations in which the key factor is virtualization of data as it allow the user to manage, handle and compute a large number of tasks very easily. By referring to Cloud technologies we mean runtime such as Hadoop, Dryad and other Map Reduce frameworks. In this paper we would analyse the above mentioned software’s and techniques for the cloud system by comparing them on the basis of its processing speed, its data handling capacity, the nature of user friendliness. We would discuss large scale data analysis using different implementations on the above mentioned tools and after that we would give a performance analysis of these tools on the given implementation like Cap3, HEP, Cloudburst.

Purpose The purpose of this paper is to evaluate the impact of scientific data in order to assess the reliability of data to support data curation, to establish trust between researchers to support reuse of digital data and encourage researchers to share more data. Design/methodology/approach The authors compared the correlations between usage counts of associated data in Dryad and citation counts of articles in Web of Science in different subject areas in order to assess the possibility of using altmetric indicators to evaluate scientific data. Findings There are high positive correlations between usage counts of data and citation counts of associated articles. The citation counts of article’s shared data are higher than the average citation counts in most of the subject areas examined by the authors. Practical implications The paper suggests that usage counts of data could be potentially used to evaluate scholarly impact of scientific data, especially for those subject areas without special data repositories. Originality/value The study examines the possibility to use usage counts to evaluate the impact of scientific data in a generic repository Dryad by different subject categories.

We present The Odonate Phenotypic Database (OPD): an online data resource of dragonfly and damselfly phenotypes (Insecta: Odonata). Odonata is a relatively small insect order that currently consists of about 6400 species belonging to 32 families. The database consists of multiple morphological, life-history and behavioral traits, and biogeographical information collected from literature sources. We see taxon-specific phenotypic databases from Odonata and other organismal groups as becoming an increasing valuable resource in comparative studies. Our database has phenotypic records for 1011 of all 6400 known odonate species. The database is accessible at http://www.odonatephenotypicdatabase.org/, and a static version with an information file about the variables in the database is archived at Dryad., Measurement(s)Invertebrate Taxonomy • body length • wing • appendage morphology trait • mating behavior • flight behavior • habitat • ecozone • climate • sexual dimorphism • size • color • biological pigmentTechnology Type(s)digital curationFactor Type(s)phenotypesSample Characteristic - OrganismZygoptera • AnisopteraSample Characteristic - Environmentstream • pond • wetland ecosystem • lake • river • ephemeral springSample Characteristic - LocationSouth America • North America • Africa • Asia • Europe • Australia Machine-accessible metadata file describing the reported data: 10.6084/m9.figshare.10321595

As cluster computing frameworks such as Spark, Dryad, Flink, and Ray are being deployed in mission critical applications and on larger and larger clusters, their ability to tolerate failures is growing in importance. These frameworks employ two broad approaches for fault tolerance: checkpointing and lineage. Checkpointing exhibits low overhead during normal operation but high overhead during recovery, while lineage-based solutions make the opposite tradeoff. We propose the lineage stash, a decentralized causal logging technique that significantly reduces the runtime overhead of lineage-based approaches without impacting recovery efficiency. With the lineage stash, instead of recording the task's information before the task is executed, we record it asynchronously and forward the lineage along with the task. This makes it possible to support large-scale, low-latency (millisecond-level) data processing applications with low runtime and recovery overheads. Experimental results for applications in distributed training and stream processing show that the lineage stash provides task execution latencies similar to checkpointing alone, while incurring a recovery overhead as low as traditional lineage-based approaches.

Achieving a high-quality reconstruction of a phylogenetic tree with branch lengths proportional to absolute time (chronogram) is a difficult and time-consuming task. But the increased availability of fossil and molecular data, and time-efficient analytical techniques has resulted in many recent publications of large chronograms for a large number and wide diversity of organisms. Knowledge of the evolutionary time frame of organisms is key for research in the natural sciences. It also represent valuable information for education, science communication, and policy decisions. When chronograms are shared in public and open databases, this wealth of expertly-curated and peer-reviewed data on evolutionary timeframe is exposed in a programatic and reusable way, as intensive and localized efforts have improved data sharing practices, as well as incentivizited open science in biology. Here we present DateLife, a service implemented as an R package and an R Shiny website application available at www.datelife.org, that provides functionalities for efficient and easy finding, summary, reuse, and reanalysis of expert, peer-reviewed, public data on time frame of evolution. The main DateLife workflow constructs a chronogram for any given combination of taxon names by searching a local chronogram database constructed and curated from the Open Tree of Life Phylesystem phylogenetic database, which incorporates phylogenetic data from the TreeBASE database as well. We implement and test methods for summarizing time data from multiple source chronograms using supertree and congruification algorithms, and using age data extracted from source chronograms as secondary calibration points to add branch lengths proportional to absolute time to a tree topology. DateLife will be useful to increase awareness of the existing variation in alternative hypothesis of evolutionary time for the same organisms, and can foster exploration of the effect of alternative evolutionary timing hypotheses on the results of downstream analyses, providing a framework for a more informed interpretation of evolutionary results.

With the advancement of ever-growing online services, distributed Big Data storage i.e. Hadoop, Dryad gained much more attention than ever and the fundamental requirements like fault tolerance and data availability become the concern for these platforms. Data replication policies in Big Data applications are shifting towards dynamic approaches based on the popularity of files. Formulation of dynamic replication factor paved the way of solving the issues generated by existing data contention in hotspots and ensuring timely data availability. But from the empirical observations, it can be deduced that popularity of files is temporal rather than perpetual in nature and, after a certain period, content's popularity ceases most of the time which introduces the I/O bottleneck of updating replication in the disk. To handle such temporal skewed popularity of contents, we propose a dynamic data replication toolset using the power of in-memory processing by integrating MemCached server into Hadoop for getting improved performance. We compare the proposed algorithm with the traditional infrastructure and vanilla memory algorithm, as the evidence from the experimental results, the proposed design performs better i.e throughput and execution period.

Stakeholder cooperation can be vital in managing conservation conflicts. Laboratory experiments show cooperation is less likely in the presence of uncertainty. Much less is known about how stakeholders in real‐life conservation conflicts respond to different types of uncertainty.We tested the effects of different sources of uncertainty on cooperative behaviour using a framed field experiment and interviews. The experiment compared a baseline scenario of perfect certainty with scenarios including either: i) scientific uncertainty about the effectiveness of a conflict‐reduction intervention; ii) administrative uncertainty about intervention funding; or iii) political uncertainty about the extent of community support. We applied these scenarios to a conservation conflict in the Outer Hebrides, Scotland, involving the management of geese to simultaneously meet both conservation and farming objectives. We asked 149 crofters (small scale farmers) if they would commit to cooperate with others by helping fund a goose management plan given the three sources of uncertainty.On average, intention to cooperate was highest (99%) in scenarios without uncertainty, and lowest under administrative uncertainty (77%). Scientific uncertainty and political uncertainty both had less of an effect, with over 95% of crofters predicted to be willing to cooperate in these scenarios. Crofters who indicated concern for other crofters suffering the impact of geese were more likely to cooperate. The longer an individual had been a crofter, the less likely they were to cooperate.Synthesis and applications. Crofters’ intention to cooperate is high but lessened by uncertainty, especially over the commitment from other stakeholders such as government, to cooperate on goose management. Existing cooperation on goose management may be at risk if uncertainty isn't reduced outright or commitments between parties are not strengthened. This has wide applicability, supporting the need for researchers and government advisers to: i) determine how uncertainty will impact intention of stakeholders to cooperate; and ii) take steps (such as uncertainty reduction, communication, or acceptance) to reduce the negative impact of uncertainty on cooperation.

e15609 Background: Colorectal cancer (CRC) is one of the most diagnosed cancers and the second leading cause of cancer-related death worldwide. Globally, more than 1.8 million people are diagnosed of colorectal cancer (CRC) in 2018. In advanced CRC patients, bevacizumab plus 5-fluorouracil-based or platinum-based therapy has become one of the standard first-line chemotherapy regimen. A recent study also found that antibiotic exposure could be inversely associated with the mortality in metastatic colorectal cancer (mCRC) patients treated with bevacizumab. However, subgroup analysis of this study was no sufficient. Therefore, we conducted a secondary analysis based on the data of this study from Dryad database. Methods: In this retrospective cohort study, 147 mCRC patients treated with bevacizumab were included. All data was obtained from Dryad database (https://doi.org/10.5061/dryad.ft5sk66). Patients were divided into follow subgroups: (1) left colon vs right colon; (2) BMI < 18.5 vs BMI 18.5-24 vs BMI≥24;(3) Age < 45 vs Age≥45, (4) no surgery vs palliative surgery vs radical surgery, (5) bevacizumab plus FOLFIRI vs bevacizumab plus capeOX/FOLFOX vs bevacizumab plus other chemotherapy regimen. Specific survival of each subgroup was analyzed through the Kaplan-Meier curve, and the survival curves of the variables were compared using the log-rank test. Results: Survival analysis found no statistically significant differences of the cumulative survival rates between left colon cancer and right colon cancer, (58.9% vs 62.5%, respectively; χ2 = 0.043, P = 0.836), groups of different BMI (57.9% of BMI < 18.5 vs 55.2% of 18.5≤BMI < 24 vs 70.7% of BMI ≥24，χ2 = 3.026, P = 0.220), Age < 45 group and Age≥45 group (54.8% vs 61.2%, respectively; χ2 = 0.001, P = 0.976), surgical treatments (No surgery 59.0% vs Palliative surgery 57.6% vs Radical Surgery 61.3%; χ2 = 1.885, P = 0.390), as well as groups of different chemotherapy regimens (66.7% of FLOFIRI vs 52.0% of CapeOX/FOLFOX vs 59.9% of others; χ2 = 1.572, P = 0.456). Conclusions: Based on the data from this study, we found that different sites of colon cancer, age, BMI, different surgical treatments, and different chemotherapy regimens did not affect the survival outcome of patients with mCRC treated with bevacizumab after antibiotic exposure. Subsequent studies with larger sample size are still needed to further elaborate the effects of different antibiotics on survival outcomes.

Purpose Data sharing is widely thought to help research quality and efficiency. Data sharing mandates are increasingly being adopted by journals and the purpose of this paper is to assess whether they work. Design/methodology/approach This study examines two evolutionary biology journals, Evolution and Heredity, that have data sharing mandates and make extensive use of Dryad. It uses a quantitative analysis of presence in Dryad, downloads and citations. Findings Within both journals, data sharing seems to be complete, showing that the mandates work on a technical level. Low correlations (0.15-0.18) between data downloads and article citation counts for articles published in 2012 within these journals indicate a weak relationship between data sharing and research impact. An average of 40-55 data downloads per article after a few years suggests that some use is found for shared life sciences data. Research limitations/implications The value of shared data uses is unclear. Practical implications Data sharing mandates should be encouraged as an effective strategy. Originality/value This is the first analysis of the effectiveness of data sharing mandates.

Purpose – In recent years, a large number of data repositories have been built and used. However, the extent to which scientific data are re-used in academic publications is still unknown. The purpose of this paper is to explore the functions of re-used scientific data in scholarly publication in different fields. Design/methodology/approach – To address these questions, the authors identified 827 publications citing resources in the Dryad Digital Repository indexed by Scopus from 2010 to 2015. Findings – The results show that: the number of citations to scientific data increases sharply over the years, but mainly from data-intensive disciplines, such as agricultural, biology science, environment science and medicine; the majority of citations are from the originating articles; and researchers tend to reuse data produced by their own research groups. Research limitations/implications – Dryad data may be re-used without being formally cited. Originality/value – The conservatism in data sharing suggests that more should be done to encourage researchers to re-use other’s data.

By abstracting away the complexity of distributed systems, large-scale data processing platforms—MapReduce, Hadoop, Spark, Dryad, etc.—have provided developers with simple means for harnessing the power of the cloud. In this paper, we ask whether we can automatically synthesize MapReduce-style distributed programs from input–output examples. Our ultimate goal is to enable end users to specify large-scale data analyses through the simple interface of examples. We thus present a new algorithm and tool for synthesizing programs composed of efficient data-parallel operations that can execute on cloud computing infrastructure. We evaluate our tool on a range of real-world big-data analysis tasks and general computations. Our results demonstrate the efficiency of our approach and the small number of examples it requires to synthesize correct, scalable programs.

Entrevista concedida à Revista Brasileira de Preservação Digital por meio do editor Gildenir Carolino Santos, também bibliotecário e pesquisador do Grupo de Pesquisa e Estudos sobre Preservação Digital – Dríade, via e-mail em dezembro de 2020.

With the increased adoption of the FAIR Principles, a wide range of stakeholders, from scientists to publishers, funding agencies and policy makers, are seeking ways to transparently evaluate resource FAIRness. We describe the FAIR Evaluator, a software infrastructure to register and execute tests of compliance with the recently published FAIR Metrics. The Evaluator enables digital resources to be assessed objectively and transparently. We illustrate its application to three widely used generalist repositories - Dataverse, Dryad, and Zenodo - and report their feedback. Evaluations allow communities to select relevant Metric subsets to deliver FAIRness measurements in diverse and specialized applications. Evaluations are executed in a semi-automated manner through Web Forms filled-in by a user, or through a JSON-based API. A comparison of manual vs automated evaluation reveals that automated evaluations are generally stricter, resulting in lower, though more accurate, FAIRness scores. Finally, we highlight the need for enhanced infrastructure such as standards registries, like FAIRsharing, as well as additional community involvement in domain-specific data infrastructure creation.

ACKNOWLEDGMENTS We thank the Tsawout and Tseycum First Nation bands for access to Mandarte and everyone who contributed to the long-term data collection. We thank the European Research Council for funding and the University of Aberdeen for generous access to the Maxwell High Performance Computing cluster. Pierre de Villemereuil, Michael B. Morrissey, and Jarrod D. Hadfield provided enlightening discussions during manuscript preparation. Joel McGlothlin and two anonymous reviewers provided further helpful comments. DATA ARCHIVING Data have been archived in the Dryad Digital Repository: https://doi.org/10.5061/dryad.p7p1jb3 (Wolak et al. 2018).

This work was supported by a NERC Doctoral Training Grant studentship to RAB. Data archived in Dryad Digital Repository at doi:10.5061/dryad.t0877 The extent to which sperm or ejaculate-derived products from different males interact during sperm competition—from kamikaze sperm to sperm incapacitation—remains controversial. Repeated matings in the parasitoid wasp Nasonia vitripennis lead to a short-term reduction of efficient sperm use by females, which is crucial for a haplodiploid organism when needing to allocate sex adaptively (i.e. by fertilizing eggs to produce daughters). Repeated matings by females in this species, therefore, constrain sex allocation through this “sperm-blocking” effect, eliciting a cost to polyandry. Here, we explore the causes and consequences of sperm blocking and test the hypothesis that it is an ejaculate-related trait associated with sperm competition. First, we show that sperm blocking, which leads to an overproduction of sons, is not correlated with success in either offensive or defensive roles in sperm competition. Then, we show that the extent of sperm blocking is not affected by self–self or kin–kin ejaculate interactions when compared to self vs nonself or kin versus nonkin sperm competition. Our results suggest that sperm blocking is not a sperm competition adaptation, but is instead associated with the mechanics of processing sperm in this species, which are likely shaped by selection on female reproductive morphology for adaptive sex allocation. Postprint

One hypothesis explaining extra-pair reproduction is that socially monogamous females mate with extra-pair males to adjust the coefficient of inbreeding (f) of extra-pair offspring (EPO) relative to that of within-pair offspring (WPO) they would produce with their socially paired male. Such adjustment of offspring f requires non-random extra-pair reproduction with respect to relatedness, which is in turn often assumed to require some mechanism of explicit pre-copulatory or post-copulatory kin discrimination. We propose three demographic processes that could potentially cause mean f to differ between individual females’ EPO and WPO given random extra-pair reproduction with available males without necessarily requiring explicit kin discrimination. Specifically, such a difference could arise if social pairings formed non-randomly with respect to relatedness or persisted non-randomly with respect to relatedness, or if the distribution of relatedness between females and their sets of potential mates changed during the period through which social pairings persisted. We used comprehensive pedigree and pairing data from free-living song sparrows (Melospiza melodia) to quantify these three processes and hence investigate how individual females could adjust mean offspring f through instantaneously random extra-pair reproduction. Female song sparrows tended to form social pairings with unrelated or distantly related males slightly less frequently than expected given random pairing within the defined set of available males. Furthermore, social pairings between more closely related mates tended to be more likely to persist across years than social pairings between less closely related mates. However, these effects were small and the mean relatedness between females and their sets of potential extra-pair males did not change substantially across the years through which social pairings persisted. Our framework and analyses illustrate how demographic and social structuring within populations might allow females to adjust mean f of offspring through random extra-pair reproduction without necessarily requiring explicit kin discrimination, implying that adjustment of offspring f might be an inevitable consequence of extra-pair reproduction. New theoretical and empirical studies are required to explore the general magnitude of such effects and quantify the degree to which they could facilitate or constrain long-term evolution of extra-pair reproduction.

In order to better understand the factors that most influence where researchers deposit their data when they have a choice, we collected survey data from researchers who deposited phylogenetic data in either the TreeBASE or Dryad data repositories. Respondents were asked to rank the relative importance of eight possible factors. We found that factors differed in importance for both TreeBASE and Dryad, and that the rankings differed subtly but significantly between TreeBASE and Dryad users. On average, TreeBASE users ranked the domain specialization of the repository highest, while Dryad users ranked as equal highest their trust in the persistence of the repository and the ease of its data submission process. Interestingly, respondents (particularly Dryad users) were strongly divided as to whether being directed to choose a particular repository by a journal policy or funding agency was among the most or least important factors. Some users reported depositing their data in multiple repositories and archiving their data voluntarily.

We thank the Robert Bosch Foundation, the Max Planck Society, and the University of St Andrews for funding. The data files from which densities of Maxwell's duikers were estimated using Distance software, and data describing start times of videos of Maxwell's duikers, have been archived at the Dryad data repository (https://doi.org/10.5061/dryad.b4c70) (Howe et al. 2017). 1. Reliable estimates of animal density and abundance are essential for effective wildlife conservation and management. Camera trapping has proven efficient for sampling multiple species, but statistical estimators of density from camera trapping data for species that cannot be individually identified are still in development. 2. We extend point-transect methods for estimating animal density to accommodate data from camera traps, allowing researchers to exploit existing distance sampling theory and software for designing studies and analyzing data. We tested it by simulation, and used it to estimate densities of Maxwell’s duikers (Philantomba maxwellii) in Taï National Park, Côte d’Ivoire. 3. Densities estimated from simulated data were unbiased when we assumed animals were not available for detection during long periods of rest. Estimated duiker densities were higher than recent estimates from line transect surveys, which are believed to underestimate densities of forest ungulates. 4. We expect these methods to provide an effective means to estimate animal density from camera trapping data and to be applicable in a variety of settings. Postprint

Apache Hadoop is an open source big data processing framework. Hadoop helps to Store and process big data. YARN efficiently share resources among the nodes in cluster. One of the key building blocks of any heterogeneous cluster computing is resource fairness. Several researchers propose models to improve efficiency of the resource utilization by sharing resources fairly among the nodes. Memory less Resource Fairness (MLRF) widely used in existing frameworks like YARN, Mesos and Dryad. Long Term Resource Fairness (LTRF) improves Resources Fairness but not up to the mark. This paper propose Combined Regression Ranking (CRRYARN). It predict ranking of different workloads based on amount resource utilization factors like CPU Intensive, Memory intensive etc., Based on this ranking the resource sharing happened among the workload in Single Tenants of the same cloud. It works on loan lending with interest free, rank based discount share model. It improves performance of the existing model by 30% effectively.

Distributed computing accomplished broad appropriation because of consequently parallelizing and transparently executing tasks in distributed environments. Straggling tasks is an essential test confronted by all Big Data Processing Frameworks for example MapReduce, Dryad, and Spark. Stragglers are the assignments that run much slower than different tasks and since a job completes just when it’s last undertaking completions, stragglers postponement work fruition. Stragglers extraordinarily impact little occupations such that employments comprising of a couple of undertakings. Such occupations are fundamentally deferred regardless of the fact that a solitary undertaking is moderate .This paper survey stragglers recognition and rescheduling systems proposed so far and brings up their strengths and shortcomings. This paper additionally displays wise attributes and impediments of the existing state- of-the- craftsmanship calculations to take care of the issue of stragglers relief.

Purpose – The purpose of this paper is to examine the effect of the Helping Interdisciplinary Vocabulary Engineering (HIVE) system on the inter-indexer consistency of information professionals when assigning keywords to a scientific abstract. This study examined first, the inter-indexer consistency of potential HIVE users; second, the impact HIVE had on consistency; and third, challenges associated with using HIVE. Design/methodology/approach – A within-subjects quasi-experimental research design was used for this study. Data were collected using a task-scenario based questionnaire. Analysis was performed on consistency results using Hooper's and Rolling's inter-indexer consistency measures. A series of t-tests was used to judge the significance between consistency measure results. Findings – Results suggest that HIVE improves inter-indexing consistency. Working with HIVE increased consistency rates by 22 percent (Rolling's) and 25 percent (Hooper's) when selecting relevant terms from all vocabularies. A statistically significant difference exists between the assignment of free-text keywords and machine-aided keywords. Issues with homographs, disambiguation, vocabulary choice, and document structure were all identified as potential challenges. Research limitations/implications – Research limitations for this study can be found in the small number of vocabularies used for the study. Future research will include implementing HIVE into the Dryad Repository and studying its application in a repository system. Originality/value – This paper showcases several features used in HIVE system. By using traditional consistency measures to evaluate a semantic web technology, this paper emphasizes the link between traditional indexing and next generation machine-aided indexing (MAI) tools.

To associate research data to the published results favors their reuse by the scientific community, but this does not afford sufficient guarantees of preservation. To store them in databases solves this contingency and provides visibility, but in Spain there are not many services of this kind. For this reason, we describe, evaluate and discuss the pros and cons of foreign multidisciplinary data repositories that can be useful for researchers and information managers: Dryad, Figshare, Zenodo and Dataverse. It is still early to choose optimally and definitively one or the other application, so we conclude with recommendations to guide the user community and intermediaries.

SUMMARY It is difficult to express the parallelism present in complex computations by using existing higher level abstractions such as MapReduce and Dryad. These computations include applications from wide variety of domains, like Artificial Intelligence, Decision Tree Algorithms, Association Rule Mining, Recommender Systems, Graph Algorithms, Clustering Algorithms, Compute Intensive Scientific Workflows, Optimization Algorithms, and so forth. Their execution graphs introduce new challenges in terms of programmer expressibility and runtime performance such as iterative and recursive computations, shared communication model, and so forth. We propose an extension to MapReduce, called Generate-Map-Reduce (GMR), targeted towards modeling these applications. GMR introduces a new Generate abstraction into the MapReduce framework that captures recursive computations. The runtime also supports iterative jobs and a distributed communication model by using shared data structures. We illustrate recursive computations with GMR by modeling complex applications such as simulated annealing, A* search, and adaptive quadrature computation that require recursive spawning of new tasks to handle variable degree of parallelism. GMR runtime supports caching of common data across iterations in memory and local disks. We illustrate how this caching helps in achieving significant speedup for iterative computations by modeling k-means clustering. Copyright © 2013 John Wiley & Sons, Ltd.

By abstracting away the complexity of distributed systems, large-scale data processing platforms—MapReduce, Hadoop, Spark, Dryad, etc.—have provided developers with simple means for harnessing the power of the cloud. In this paper, we ask whether we can automatically synthesize MapReduce-style distributed programs from input–output examples. Our ultimate goal is to enable end users to specify large-scale data analyses through the simple interface of examples. We thus present a new algorithm and tool for synthesizing programs composed of efficient data-parallel operations that can execute on cloud computing infrastructure. We evaluate our tool on a range of real-world big-data analysis tasks and general computations. Our results demonstrate the efficiency of our approach and the small number of examples it requires to synthesize correct, scalable programs.

It was recently proposed that long-term population studies be exempted from the expectation that authors publicly archive the primary data underlying published articles. Such studies are valuable to many areas of ecological and evolutionary biological research, and multiple risks to their viability were anticipated as a result of public data archiving (PDA), ultimately all stemming from independent reuse of archived data. However, empirical assessment was missing, making it difficult to determine whether such fears are realistic. I addressed this by surveying data packages from long-term population studies archived in the Dryad Digital Repository. I found no evidence that PDA results in reuse of data by independent parties, suggesting the purported costs of PDA for long-term population studies have been overstated., Ecologists were recently advised that primary data archiving threatens the sustainability of long-term population studies because of negative impacts of data reuse by others. However, this Perspective describes an empirical assessment that finds little evidence of data reuse, suggesting that such fears may be unfounded.

Executing large number of self-regulating tasks or tasks that execute minimal inter-task communication in analogous is a common requirement in many domains. In this paper, we present our knowledge in applying two new Microsoft technologies Dryad and Azure to three bioinformatics applications. We also contrast with traditional MPI and Apache Hadoop MapReduce completion in one example. The applications are an EST (Expressed Sequence Tag) series assembly program, PhyloD statistical package to recognize HLA-associated viral evolution, and a pairwise Alu gene alignment application. We give detailed presentation discussion on a 768 core Windows HPC Server cluster and an Azure cloud. All the applications start with a “doubly data parallel step” connecting independent data chosen from two parallel (EST, Alu) or two different databases (PhyloD). There are different structures for final stages in each application.

With the unprecedented growth of data generated by mankind nowadays, it has become critical to develop efficient techniques for processing these massive data sets. To tackle such challenges, analytical data processing systems must be extremely efficient, scalable, and flexible as well as economically effective. Recently, Hadoop, an open-source implementation of MapReduce, has gained interests as a promising big data processing system. Although Hadoop offers the desired flexibility and scalability, its performance has been noted to be suboptimal when it is used to process complex analytical tasks. This paper presents E3, an elastic and efficient execution engine for scalable data processing. E3 adopts a “middle” approach between MapReduce and Dryad in that E3 has a simpler communication model than Dryad yet it can support multi-stages job better than MapReduce. E3 avoids reprocessing intermediate results by adopting a stage-based evaluation strategy and collocating data and user-defined (map or reduce) functions into independent processing units for parallel execution. Furthermore, E3 supports block-level indexes, and built-in functions for specifying and optimizing data processing flows. Benchmarking on an in-house cluster shows that E3 achieves significantly better performance than Hadoop, or put it another way, building an elastically scalable and efficient data processing system is possible.

A data-parallel framework is very attractive for large-scale data processing since it enables such an application to easily process a huge amount of data on commodity machines. MapReduce, a popular data-parallel framework, is used in various fields such as web search, data mining and data warehouses; it is proven to be very practical for such a data-parallel application. A star-join query is a popular query in data warehouses that are a current target domain of data-parallel frameworks. This article proposes a new algorithm that efficiently processes star-join queries in data-parallel frameworks such as MapReduce and Dryad. Our star-join algorithm for general data-parallel frameworks is called Scatter-Gather-Merge, and it processes star-join queries in a constant number of computation steps, although the number of participating dimension tables increases. By adopting bloom filters, Scatter-Gather-Merge reduces a non-trivial amount of IO. We also show that Scatter-Gather-Merge can be easily applied to MapReduce. Our experimental results in both cluster and cloud environments show that Scatter-Gather-Merge outperforms existing approaches.

The growing demand for large-scale data mining and data analysis applications has led both industry and academia to design new types of highly scalable data-intensive computing platforms. MapReduce and Dryad are two popular platforms in which the dataflow takes the form of a directed acyclic graph of operators. These platforms lack built-in support for iterative programs, which arise naturally in many applications including data mining, web ranking, graph analysis, model fitting, and so on. This paper presents HaLoop, a modified version of the Hadoop MapReduce framework that is designed to serve these applications. HaLoop not only extends MapReduce with programming support for iterative applications, it also dramatically improves their efficiency by making the task scheduler loop-aware and by adding various caching mechanisms. We evaluated HaLoop on real queries and real datasets. Compared with Hadoop, on average, HaLoop reduces query runtimes by 1.85, and shuffles only 4% of the data between mappers and reducers.

The Dryad repository is for data supporting published research in the field of evolutionary biology and related disciplines. Dryad development team members seek a theoretical framework to aid communication about metadata issues and plans. This article explores lifecycle modeling as a theoretical framework for understanding metadata in the repository environment. A background discussion reviews the importance of theory, the status of a metadata theory, and lifecycle concepts. An analysis draws examples from the Dryad repository demonstrating automatic propagation, metadata inheritance, and value system adoption, and reports results from a faceted term mapping experiment that included 12 vocabularies and approximately 600 terms. The article also reports selected key findings from a recent survey on the data-sharing attitudes and behaviors of nearly 400 evolutionary biologists. The results confirm the applicability of lifecycle modeling to Dryad's metadata infrastructure. The article concludes that lifecycle...

The operative term in this quote from the late provocateur/ philosopher of science is “capable.” While all ideas and data are capable of improving knowledge, there is no guarantee that they will actually do so. From the beginning of time the crucial first step was being aware of the ideas. A century ago this was limited by the ability to gain access to books, journals, and colleagues. Overcoming these limitations was straightforward; gain access to the published work and interact with others in the field. Subsequently possessing the ability to overcome one’s own biases and parochial worldviews to recognize that the ideas of others have value is an entirely different matter, but the point here is that we must first be able to access this information. The data explosion that we are witnessing in science can be considered a triumph of human ingenuity or a tragedy of futility. The generation of new information “should” be viewed positively. Data derived from well-designed experiments has great potential to advance knowledge. However, if we do not aggressively work to build the framework to organize and exchange this information then it will be for naught. Indeed, it could well lead to the opposite of the desired effect; an avalanche of useless information that will just stifle the scientific enterprise. We can no longer rely on library access and scientific conferences to provide access to useful scientific information.

Adult individuals that do not breed in a given year occur in a wide range of natural populations. However, such nonbreeders are often ignored in theoretical and empirical population studies, limiting our knowledge of how nonbreeders affect realized and estimated population dynamics and potentially impeding projection of deterministic and stochastic population growth rates. We present and analyse a general modelling framework for systems where breeders and nonbreeders differ in key demographic rates, incorporating different forms of nonbreeding, different life histories and frequency-dependent effects of nonbreeders on demographic rates of breeders. Comparisons of estimates of deterministic population growth rate, λ, and demographic variance, σd2, from models with and without distinct nonbreeder classes show that models that do not explicitly incorporate nonbreeders give upwardly biased estimates of σd2, particularly when the equilibrium ratio of nonbreeders to breeders, Nnb∗/Nb∗, is high. Estimates of λ from empirical observations of breeders only are substantially inflated when individuals frequently re-enter the breeding population after periods of nonbreeding. Sensitivity analyses of diverse parameterizations of our model framework, with and without negative frequency-dependent effects of nonbreeders on breeder demographic rates, show how changes in demographic rates of breeders vs. nonbreeders differentially affect λ. In particular, λ is most sensitive to nonbreeder parameters in long-lived species, when Nnb∗/Nb∗0, and when individuals are unlikely to breed at several consecutive time steps. Our results demonstrate that failing to account for nonbreeders in population studies can obscure low population growth rates that should cause management concern. Quantifying the size and demography of the nonbreeding section of populations and modelling appropriate demographic structuring is therefore essential to evaluate nonbreeders' influence on deterministic and stochastic population dynamics.

A diabetes patient web portal allows patients to access their personal health record and may improve diabetes outcomes; however, patients' adoption is slow. We aimed to get insight into patients' experiences with a web portal to understand how the portal is being used, how patients perceive the content of the portal and to assess whether redesign of the portal might be needed.A survey among 1500 patients with type 1 and type 2 diabetes with a login to a patient portal.62 primary care practices and one outpatient hospital clinic, using a combined patient portal. We compared patients who requested a login but never used it or once ('early quitters') with patients who used it at least two times ('persistent users').632 patients (42.1%) returned the questionnaire. Their mean age was 59.7 years, 63.1% was male and 81.8% had type 2 diabetes. 413 (65.3%) people were persistent users and 34.7% early quitters. In the multivariable analysis, insulin use (OR2.07; 95%CI[1.18-3.62]), experiencing more frequently hyperglycemic episodes (OR1.30;95%CI[1.14-1.49]) and better diabetes knowledge (OR1.02, 95%CI[1.01-1.03]) do increase the odds of being a persistent user. Persistent users perceived the usefulness of the patient portal significantly more favorable. However, they also more decisively declared that the patient portal is not helpful in supporting life style changes. Early quitters felt significantly more items not applicable in their situation compared to persistent users. Both persistent users (69.8%) and early quitters (58.8%) would prefer a reminder function for scheduled visits. About 60% of both groups wanted information about medication and side-effects in their portal.The diabetes patient web portal might be improved significantly by taking into account the patients' experiences and attitudes. We propose creating separate portals for patients on insulin or not.

Data-parallel computing frameworks (DCF) such as MapReduce, Spark, and Dryad etc. Have tremendous applications in big data and cloud computing, and throw tons of flows into data center networks. In this paper, we design and implement FLOW PROPHET, a general framework to predict traffic flows for DCFs. To this end, we analyze and summarize the common features of popular DCFs, and gain a key insight: since application logic in DCFs is naturally expressed by directed acyclic graphs (DAG), DAG contains necessary time and data dependencies for accurate flow prediction. Based on the insight, FLOW PROPHET extracts DAGs from user applications, and uses the time and data dependencies to calculate flow information 4-tuple, (source, destination, flow size, establish time), ahead-of-time for all flows. We also provide generic programming interface to FLOW PROPHET, so that current and future DCFs can deploy FLOW PROPHET readily. We implement FLOW PROPHET on both Spark and Hadoop, and perform extensive evaluations on a testbed with 37 physical servers. Our implementation and experiments demonstrate that, with time in advance and minimal cost, FLOW PROPHET can achieve almost 100% accuracy in source, destination, and flow size predictions. With accurate prediction from FLOW PROPHET, the job completion time of a Hadoop TeraSort benchmark is reduced by 12.52% on our cluster with a simple network scheduler.

Poor quality metadata can have negative impact not only on the way research datasets are retrieved, shared and used by scientists, but also on the way research data repositories are managed and audited. The aim of the research reported in this paper was to perform a descriptive analysis of the Dublin Core's Subject metadata element and identify its quality problems, if any, in the context of the Dryad research data repository following a novel data-preprocessing method using SQL queries. The findings showed quality problems related to the lack of controlled vocabulary and standardisation, like the inconsistent use of singular and plural forms, adjectives and synonyms. This study has both practical and methodological implications for the evaluation of metadata and the improvement of the quality of the research data annotation process in open research data repositories.

Studies of chimpanzee vocal communication provide valuable insights into the evolution of communication in complex societies, and also comparative data for understanding the evolution of human language. One particularly valuable dataset of recordings from free-living chimpanzees was collected by Frans X. Plooij and the late Hetty van de Rijt-Plooij at Gombe National Park, Tanzania (1971–73). These audio specimens, which have not yet been analysed, total over 10 h on 28 tapes, including 7 tapes focusing on adult individuals with a total of 605 recordings. In 2014 the first part of that collection of audio specimens covering the vocalizations of the immature Gombe chimpanzees was made available. The data package described here covers the vocalizations of the adult chimpanzees. We expect these recordings will prove useful for studies on topics including referential signalling and the emergence of dialects. The digitized sound recordings were stored in the Macaulay Library and the Dryad Repository. In addition, the original notes on the contexts of the calls were translated and transcribed from Dutch into English. Machine-accessible metadata file describing the reported data (ISA-Tab format)

Many of the historic turning points in the history of evolutionary science are examples of ‘synthetic research’, in which new knowledge was generated through the integration of existing data, methods, results and concepts1. This tradition goes back to Darwin’s famously multifaceted case for evolution by natural selection in The Origin of Species, the reconciliation of Mendelian and statistical genetics by R.A. Fisher2, and the Modern Evolutionary Synthesis of the mid-20th century that brought together population genetics, paleontology and other schools of evolutionary theory3. For the past decade, with funding from the US National Science Foundation (NSF), National Evolutionary Synthesis Center (NESCent) has promoted the continuation of this tradition through competitive support for researchers who wish to apply synthesis to any area of evolutionary science of their choosing. While the center runs several programs, the ones directly responsible for the most research outputs are the resident scholars (graduate, postdoctoral and sabbatical fellows) and working groups (with recurrent face to face meetings over two years with participants from a diversity of disciplines and from institutions around the world). These programs provide researchers with opportunities to pursue ideas for synthetic research that are much more difficult to fund through traditional channels4. Data are both a critical input and output for many synthetic investigations. Projects supported by the center often invest a great deal of effort in the collection, curation, and integration of existing data for meta-analysis, even though, as a matter of policy, NESCent does not support the collection of new data. In other cases, worthwhile research projects have had to be declined at the proposal stage to due to the inaccessibility of previously collected data for reuse. In order to make it possible for future researchers to build upon the work being done today, NESCent has had a policy since 2006 that data created through NESCent-sponsored activities be open (http://nescent.org/public_documents/Informatics_Policy/Data_and_Software_Policy.pdf). More precisely, the policy stipulates that all data (and software) are to be made publicly available no later than one year after the conclusion of the NESCent award or immediately upon publication of the results, whichever comes earlier. Data should be deposited in a public data repository with no restrictions on use and dissemination beyond the form of attribution. Furthermore, the data should be adequately documented for validation and reuse, including appropriate attribution of its original source. To assist in seeing this policy carried through, NESCent has provided support ranging from consultation (e.g., metadata standards or licensing) to digitization to the provision of specialized tools for collaborative data wrangling. Recognizing that a policy that applies only to NESCent supported research would have a relatively limited impact, NESCent contributed to adoption of the Joint Data Archiving Policy (JDAP) by many key journals in evolution and ecology beginning in 2011 (http://datadryad.org/pages/jdap). An important role played by the center was incubating the Dryad Digital Repository, which now provides a trusted home for data associated with the scholarly record for which a specialized long-term repository is lacking5, including much of the data described in this Collection. The JDAP and the advent of Dryad have been responsible for a dramatic uptick in the availability of data associated with traditional publications in the field6. However, a traditional publication is not the best vehicle for dissemination when the data themselves make a standalone contribution to scholarship. This rolling Collection provides an outlet for NESCent-sponsored researchers to publish uniquely valuable data in a coherent, independently citable package, carefully described with standardized metadata. All the Data Descriptors are linked to preservation snapshots of the data in a form that will be reusable for years to come and, in some cases, these are complemented by more detailed or more dynamic data access mechanisms. The first four Data Descriptors to be published in the Collection nicely illustrate some of the diversity of data outputs generated through NESCent-sponsored activities, including literature-based data compilations, the results from long-term experimental and observational programs, and digitized historic records. One common mode of synthetic research is to compile data points that have been reported at various places in the literature into a single dataset that is amenable to comparative analysis or meta-analysis, as exemplified by the contribution from the Tree of Sex Consortium7. The 17 authors in the Consortium, participants in a NESCent Working Group, wished to understand the drivers behind the diversity of genetic and environmental sex determination systems in nature. Collectively, they had expertise in several different taxonomic groups and a number of different aspects of sex determination. This range of expertise allowed them to compile a high-quality dataset of more than 20 variables about sexual systems from tens of thousands of species across plants, vertebrates and other animals. The snapshot of data in Dryad is complemented by a living online database hosted by the consortium (http://purl.org/nescent/treeofsex). The contribution from Conner et al.8 presents the detailed results from a long-term artificial selection experiment on floral traits. It includes data that underlie existing publications together with data that have yet to be analyzed and published. A NESCent sabbatical fellowship allowed the corresponding author to compile these years of results, representing a significant fraction of his professional career, into a well-documented whole that can be built upon by others. The two other contributions in the initial collection, from Zehr et al.9 and Plooij et al.10 make available for reuse uniquely valuable observational data from primates. Zehr et al.9 report the life history data for 3,627 captive individuals from 27 different strepsirrhines (lemurs, lorises and galagos). The data are particularly valuable in combination with the large collection of associated biological samples and the live research colonies at the Duke Lemur Center. Plooij et al.10 describe the largest dataset of recordings from free-living juvenile chimpanzees, originally collected at Gombe National Park, Tanzania in the early 1970s. The original recordings of 16 animals have been digitized by the Macauley Library and annotated with detailed contextual field notes. These are now available from both the Macaulay Library and in raw form from Dryad. Such irreplaceable datasets are clearly deserving of careful documentation, preservation and stewardship.

The recent increase in interest for batch analytics has resulted in extensive use of distributed frameworks such as Hadoop and Dryad. Batch analytics-as the name suggests, perform many computations on large volumes of data. That is, large quantities of data are ingested once and read many times mostly in large chunks, which is characterized as write-once read-many (WORM) workload. The storage part of these distributed frameworks (say, HDFS in Hadoop) use file systems such as ext4 or XFS as native object stores to store objects as files in individual nodes of the distributed system. These general purpose file systems were designed with broader goals such as POSIX-compliance, optimal performance for a wide range of file size, user friendliness, etc. However, most of these features are not required for a native object store in distributed file systems. WORM Store is a light weight object store that is designed exclusively for use in distributed systems for WORM workload. WORM Store provides interesting advantages such as the ability to pre-fetch large objects, small metadata to data ratio, media aware data/metadata placement, etc. As WORM Store is log-structured, it provides the ability to recover upon failure. Our experiments show that WORM Store provides a 28% increase in the read throughput per node in a Hadoop cluster.

Data preservation has gained momentum and visibility in connection with the growth in digital data and data sharing policies. The Dryad Repository, a curated general–purpose repository for preserving and sharing the data underlying scientific publications, has taken steps to develop a preservation policy to ensure the long–term persistence of this archived data. In 2013, a Preservation Working Group, consisting of Dryad staff and national and international experts in data management and preservation, was convened to guide the development of a preservation policy. This paper describes the policy development process, outcomes, and lessons learned in the process. To meet Dryad’s specific needs, Dryad’s preservation policy negotiates between the ideal and the realistic, including complying with broader governing policies, matching current practices, and working within system constraints.

In addition to encouraging the deposit of research data into institutional data repositories, academic librarians can further support research data sharing by facilitating the deposit of data into external disciplinary data repositories. In this paper, we focus on the University of Michigan Library and Dryad, a repository for scientific and medical data, as a case study to explore possible forms of partnership between academic libraries and disciplinary data repositories. We found that although few University of Michigan researchers have submitted data to Dryad, many have recently published articles in Dryad-integrated journals, suggesting significant opportunities for Dryad use on our campus. We suggest that academic libraries could promote the sharing and preservation of science and medical data by becoming Dryad members, purchasing vouchers to cover researchers’ data submission costs, and hosting local curators who could directly work with campus researchers to improve the accuracy and completeness of data packages and thereby increase their potential for re-use. By enabling the use of both institutional and disciplinary data repositories, we argue that academic librarians can achieve greater success in capturing the vast amounts of data that presently fail to depart researchers’ hands and making that data visible to relevant communities of interest.