Skip to content

This is a repository to discuss, collect and store the requirements for scientific workflow systems

License

Notifications You must be signed in to change notification settings

montoyav/NFDI4IngScientificWorkflowRequirements

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Introduction - tools for scientific workflows

Software-driven scientific workflows are often characterized by a complex interplay of various pieces of software executed in a particular order. The output of a computational step may serve as input to a subsequent computation, which requires them to be processed sequentially with a proper mapping of outputs to inputs. Other computations are independent of each other and can be executed in parallel. Thus, one of the main tasks of a workflow tool is a proper and efficient scheduling of the individual processing steps.

Each processing step, just as the workflow itself, typically processes some input and produces output data. Apart from changing the input data to operate on, processing steps can usually be configured by a set of parameters to change their behavior. Moreover, the behavior of a processing step is determined by its source code and/or executable binaries/packages that are called within it. Beyond this, the computation environment not only has a significant influence on its behavior, but is also crucial in order for the processing step to work at all. The environment includes the versions of the interpreters or compilers, as well as all third-party libraries and packages that contribute to the computations carried out in a processing step.

In order for research workflows to comply with the FAIR principles, they should be:

  • published
  • documented
  • annotated with metadata
  • reproducible
  • reusable

Because of the dependency on their environment, reproducibility and reusability is not trivial to achieve for scientific workflows. The results presented in a scientific paper, for instance, are produced with a very specific environment, and in order to guarantee reproducibility, this environment has to be reinstantiated possibly many years later.

In this documentation we want to elaborate, together with the scientific community, a vision on how scientific workflows should be created/packaged/published in order to be as FAIR as possible. We want to evaluate if existing workflow tools provide reusable solutions and identify the capabilities that are missing to reach our goal: reproducible research workflows, by anyone, anywhere and anytime.

Go to the documentation.

About

This is a repository to discuss, collect and store the requirements for scientific workflow systems

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 61.5%
  • Common Workflow Language 22.9%
  • Nextflow 12.2%
  • TeX 2.3%
  • GLSL 1.1%