TY - CONF
T1 - Reorganizing workflow evolution provenance (USENIX Workshop on the Theory and Practice of Provenance (TaPP)
AU - Koop, David
AU - Freire, Juliana
N1 - Funding Information:
There are two extensions that we plan to explore in future work. First, while creating version trees from scratch can be a challenging problem, we would like to explore techniques that build evolution provenance from collections of workflows using heuristics like those in phylogenetic tree construction [7]. Second, we plan to consider using directed graphs (instead of trees) to create smaller provenance footprints for workflow evolution. Directed graphs present potential merging issues as seen in darcs [5] but may also better allow integration of different explorations. Acknowledgments. This work was supported in part by the National Science Foundation (CNS-1229185, IIS-1142013). References [1] Y. Amsterdamer, D. Deutch, T. Milo, and V. Tannen. On provenance minimization. In Proc. PODS 2011, pages 141–152. ACM, 2011. [2] M. K. Anand, S. Bowers, T. McPhillips, and B. Ludäscher. Efficient provenance storage over nested data collections. In Proc. EDBT 2009, pages 958–969. ACM, 2009. [3] S. Chacon. Pro Git. Apress, Berkely, CA, USA, 1st edition, 2009. http://git-scm.com/book. [4] A. P. Chapman, H. V. Jagadish, and P. Ramanan. Efficient provenance storage. In Proc. SIGMOD 2008, pages 993–1006. ACM, 2008. [5] Darcs. http://darcs.net/. [6] E. D. Demaine, S. Langerman, and E. Price. Confluently persistent tries for efficient version control. In Algorithm Theory–SWAT 2008, pages 160–172. Springer, 2008. [7] J. Felsenstein. Inferring phylogenies. Sinauer Associates Sunderland, 2nd edition, 2004. [8] J. Freire, D. Koop, E. Santos, C. Scheidegger, C. Silva, and H. T.
Publisher Copyright:
© 6th Workshop on the Theory and Practice of Provenance, TaPP 2014. All rights reserved.
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2014
Y1 - 2014
N2 - The provenance of related computations presents the opportunity to better understand and explore the differences and similarities of various approaches. As users design and refine workflows, evolution provenance captures the relationships between workflows as actions that mutate one workflow to another. However, such provenance may not always be the most compact or intuitive. This paper presents algorithms to update and transform workflow evolution provenance to achieve a representation that better exposes the correspondences between computations. We evaluate these algorithms based on the efficiency of the representation as well as the speed of the transformation.
AB - The provenance of related computations presents the opportunity to better understand and explore the differences and similarities of various approaches. As users design and refine workflows, evolution provenance captures the relationships between workflows as actions that mutate one workflow to another. However, such provenance may not always be the most compact or intuitive. This paper presents algorithms to update and transform workflow evolution provenance to achieve a representation that better exposes the correspondences between computations. We evaluate these algorithms based on the efficiency of the representation as well as the speed of the transformation.
UR - http://www.scopus.com/inward/record.url?scp=84976631008&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84976631008&partnerID=8YFLogxK
M3 - Paper
AN - SCOPUS:84976631008
T2 - 6th Workshop on the Theory and Practice of Provenance, TaPP 2014
Y2 - 12 June 2014 through 13 June 2014
ER -