TY - JOUR
T1 - The Rockerverse
T2 - Packages and Applications for Containerisation with R
AU - Nüst, Daniel
AU - Eddelbuettel, Dirk
AU - Bennett, Dom
AU - Cannoodt, Robrecht
AU - Clark, Dav
AU - Daróczi, Gergely
AU - Edmondson, Mark
AU - Fay, Colin
AU - Hughes, Ellis
AU - Kjeldgaard, Lars
AU - Lopp, Sean
AU - Marwick, Ben
AU - Nolis, Heather
AU - Nolis, Jacqueline
AU - Ooi, Hong
AU - Ram, Karthik
AU - Ross, Noam
AU - Shepherd, Lori
AU - Sólymos, Péter
AU - Swetnam, Tyson Lee
AU - Turaga, Nitesh
AU - Petegem, Charlotte Van
AU - Williams, Jason
AU - Willis, Craig
AU - Xiao, Nan
N1 - Funding Information:
Second, the Whole Tale project (https://wholetale.org) combines the strengths of the Rocker Project’s curated Docker images with repo2docker. Whole Tale is a National Science Foundation (NSF) funded project developing a scalable, open-source, multi-user platform for reproducible research (Brinckman et al., 2019; Chard et al., 2019b). A central goal of the platform is to enable researchers to easily create and publish executable research objects37 associated with published research (Chard et al., 2019a). Using Whole Tale, researchers can create and publish Rocker-based reproducible research objects to a growing number of repositories including DataONE member nodes, Zenodo and soon Dataverse. Additionally, Whole Tale supports automatic data citation and is working on capabilities for image preservation and provenance capture to improve the transparency of published computational research artefacts (Mecum et al., 2018; McPhillips et al., 2019). For R users, Whole Tale extends the Jupyter Project’s repo2docker tool to simplify the customisation of R-based environments for researchers with limited experience with either Docker or Git. Multiple options have been discussed to allow users to change the Ubuntu LTS (long-term support, currently Bionic Beaver) base image, buildpack-deps:bionic, used in repo2docker. Whole Tale implemented a custom RockerBuildPack38. The build pack combines a rocker/geospatial image with repo2docker’s composability39. This works because both Rocker images and the repo2docker base image use distributions with APT (Wikipedia contributors, 2020a) so that the instructions created by the latter work because of the compatible shell and package manager.
Funding Information:
DN is supported by the project Opening Reproducible Research (o2r) funded by the German Research Foundation (DFG) under project number PE 1632/17-1. The funders had no role in data collection and analysis, decision to publish, or preparation of the manuscript. KR was supported in part by a grant from The Leona M. and Harry B. Helmsley Charitable Trust, award number 2016PG-BRI004. LS and NT are supported by US NIH / NHGRI awards U41HG00405 and U24HG010263. CW is supported by the Whole Tale project (https://wholetale.org) funded by the US National Science Foundation (NSF) under award OAC-1541450. NR is supported in part by the Chan-Zuckerberg Initiative Essential Open Source Software for Science program. We would like to thank Celeste R. Brennecka from the Scientific Editing Service of the University of Münster for her editorial support.
Publisher Copyright:
© 2020, The R Journal Vol. All Rights Reserved
PY - 2020
Y1 - 2020
N2 - The Rocker Project provides widely used Docker images for R across different application scenarios. This article surveys downstream projects that build upon the Rocker Project images and presents the current state of R packages for managing Docker images and controlling containers. These use cases cover diverse topics such as package development, reproducible research, collaborative work, cloud-based data processing, and production deployment of services. The variety of applications demonstrates the power of the Rocker Project specifically and containerisation in general. Across the diverse ways to use containers, we identified common themes: reproducible environments, scalability and efficiency, and portability across clouds. We conclude that the current growth and diversification of use cases is likely to continue its positive impact, but see the need for consolidating the Rockerverse ecosystem of packages, developing common practices for applications, and exploring alternative containerisation software.
AB - The Rocker Project provides widely used Docker images for R across different application scenarios. This article surveys downstream projects that build upon the Rocker Project images and presents the current state of R packages for managing Docker images and controlling containers. These use cases cover diverse topics such as package development, reproducible research, collaborative work, cloud-based data processing, and production deployment of services. The variety of applications demonstrates the power of the Rocker Project specifically and containerisation in general. Across the diverse ways to use containers, we identified common themes: reproducible environments, scalability and efficiency, and portability across clouds. We conclude that the current growth and diversification of use cases is likely to continue its positive impact, but see the need for consolidating the Rockerverse ecosystem of packages, developing common practices for applications, and exploring alternative containerisation software.
UR - http://www.scopus.com/inward/record.url?scp=85097018976&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85097018976&partnerID=8YFLogxK
U2 - 10.32614/rj-2020-007
DO - 10.32614/rj-2020-007
M3 - Article
AN - SCOPUS:85097018976
SN - 2073-4859
VL - 12
SP - 437
EP - 461
JO - R Journal
JF - R Journal
IS - 1
ER -