Explore - Scientific Workflow Management Tool

버전 1.0.11 (116 KB) 작성자: Jonathan A
Lightweight workflow management tool to automatically retrieve previous intermediate results and track provenance in scientific workflows.
다운로드 수: 48
업데이트 날짜: 2019/9/3

Explore is a scientific workflow management tool. You can visualize the data provenance graph and when your code is changed and re-executed, only the necessary functions are re-executed to provide an updated version of your data. This can save a lot of time, especially when experiments are constantly refined. You can easily declare an experiment as a directed acyclic graph (DAG) where the nodes are functions and the edges represent variables that are produced and consumed by functions.

During the first execution of the graph, variables are persisted to the disk which implies a longer graph execution time (due to variable loading and saving duration). However, for future executions, if the node main function and all the sub-functions called during the node execution remain unchanged (a) and the input variables of the node also remain unchanged (b), then the results are retrieved from the disk and not re-computed.

For (a), the major time consumer is the code analysis to find out the sub-functions involved in the execution of a node. Therefore, this "node sub-function dependency" information is also persisted based on the content of the sub-function files. Then, if the sub-function files did not change, it is assumed the node's sub-function dependencies did not change.
For (b), either the variable content is hashed (i) or the variable mat-file date is taken (ii) as signature to check that the variables did not change. Option (i) is advantageous as re-computed nodes might still output the same values. In that case, unchanged child nodes will not be re-computed. However, option (ii) might be faster for bigger variables, where hashing would take too much time.

Thank you to:
- Jan (2019). DataHash (https://www.mathworks.com/matlabcentral/fileexchange/31272-datahash), MATLAB Central File Exchange. Retrieved April 15, 2019.
- Aslak Grinsted (2019). cachedcall (https://de.mathworks.com/matlabcentral/fileexchange/49949-cachedcall), MATLAB Central File Exchange. Retrieved April 15, 2019.
- Aslak Grinsted (2019). rgb.m (https://de.mathworks.com/matlabcentral/fileexchange/1805-rgb-m), MATLAB Central File Exchange. Retrieved April 15, 2019.
- All upstream dependencies

For tutorials, detailed information and bug reports, please check the Github (https://www.github.com/jahsue78/explore) or the corresponding conference paper [1].

[1] Ah Sue, Jonathan, et al. "Optimizing Exploratory Workflows for Embedded Platform Trace Analysis and Its Application to Mobile Devices." International Conference on Human-Computer Interaction. Springer, Cham, 2019.

인용 양식

Jonathan A (2024). Explore - Scientific Workflow Management Tool (https://github.com/vacoa/explore), GitHub. 검색됨 .

MATLAB 릴리스 호환 정보
개발 환경: R2017a
모든 릴리스와 호환
플랫폼 호환성
Windows macOS Linux
Help CenterMATLAB Answers에서 Graphics Object Identification에 대해 자세히 알아보기

받음: cachedcall, DataHash, rgb.m

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

GitHub 디폴트 브랜치를 사용하는 버전은 다운로드할 수 없음

버전 게시됨 릴리스 정보

Add link to the paper


Update description and add new basic tutorial.


Change title and summary


Update description


Update description and acknowledgments


New description


Modify readme and add keywords.


Update description


Add Github reference.


Change keywords


Update image


이 GitHub 애드온의 문제를 보거나 보고하려면 GitHub 리포지토리로 가십시오.
이 GitHub 애드온의 문제를 보거나 보고하려면 GitHub 리포지토리로 가십시오.