First International Workshop on Workflow Science (WoWS 2017)

Tuesday October 24, 2017, 8:30am-12:35pm

In conjunction with the Thirteenth IEEE eScience Conference in Auckland, New Zealand from 24 – 27 October 2017.


Many science communities depend on large, collocated or distributed computing infrastructures for their scientific discoveries rather than on small, single-user machines or clusters. As data analysis tasks and data volumes grow larger and more complex, the software infrastructure required to manage the data, executables, and resources are likewise growing in sophistication and capability. Scientific workflows simplify a series of complex tasks and manage a collection of infrastructure components to accomplish a scientific task. Predicting and/or explaining the performance of scientific workflows is an outstanding problem for scientists, tool developers and resource administrators. Although it is much more art than science at the moment, there are research efforts to scientifically analyze workflow behaviour and performance, that we call “workflow science” or “the science of scientific workflows”. Understanding the behavior of individual tasks and the interaction among the tasks are some of the challenges this new research area focuses on, with an ultimate goal to improve the performance of workflows and utilization of the systems surrounding these workflows. This workshop will provide a forum to advance research that will significantly enhance our ability to predict how scientific workflows perform and to explain workflow behavior on distributed infrastructure comprising of computers, storage, instruments and multi-gigabit networks. The workshop envisions the study of workflows as a science of its own with theory, experimentation and applications, and accepts papers in all these three areas.

Topics of interest include, but is not limited to:

  • Analytical and data-driven models of workflow components
  • Integrated end-to-end workflow performance models including tools, data and cyberinfrastructure
  • Workflow performance monitoring, prediction and optimization tools
  • Dynamic data-driven workflow schedulers 
  • Interaction of systems and workflows
  • Analytical models for workflows on experimental, observational and simulation facilities
  • Advanced workflow applications that span data, compute and network systems and tools (pros/cons/lessons learned)
  • Vision papers on types of science that benefit from workflow approach
  • Future of workflows, research directions and vision and advanced technologies
  • Instrumentation of workflows and infrastructure components for data collection, correlation, and aggregation

Important Dates:

Submissions Due: Friday 28 July 2017
Notification of Acceptance: Wednesday 9 August 2017


Authors are invited to submit unpublished, original work, using the IEEE 8.5 × 11 manuscript guidelines: double-column text using single-spaced 10 point font on 8.5 × 11 inch pages. Templates are available from hereThe conference proceedings will be made available online through the IEEE Digital Library. Authors should submit papers hereSubmissions will be fully peer-reviewed. It is a requirement that at least one author of each accepted paper attend the conference.


Ilkay Altintas, San Diego Supercomputer Center, UC San Diego, USA
Raj Kettimuthu, Argonne National Laboratory and The University of Chicago, USA
Craig E. Tull, Lawrence Berkeley National Laboratory, USA

Program Committee:

Moustafa AbdelBaky, Rutgers University
Shantenu Jha, Rutgers University
Daniel Katz, University of Illinois at Urbana-Champaign
Scott Klasky, Oak Ridge National Laboratory
Kerstin Kleese van Dam, Brookhaven National Laboratory
Manish Parashar, Rutgers University
Douglas Thain, University of Notre Dame
Jianwu Wang, University of Maryland, Baltimore County

Steering Committee:

Rich Carlson, DOE Office of Science, Advanced Scientific Computing Research, USA
Ewa Deelman, University of Southern California, USA
Ian Foster, University of Chicago
Darren Kerbyson, Pacific Northwest National National Laboratory, USA
Erich Strohmaier, Lawrence Berkeley National Laboratory, USA

WORKSHOP PROGRAM: Tuesday October 24th, 2017

08:30am Welcome and introduction from the chairs
08:35am Keynote: Workflow Science or Workflow Engineering? , Ian Foster
09:30am Invited Talk: Supporting Data-driven Workflows Enabled by Large Scale Observatories, Manish Parashar
10:00am Break - 30 minutes
10:30am Invited Talk: Workflows for Science and Engineering, David Abramson
11:10am Research Paper: Deep Learning on Operational Facility Data Related to Large-Scale Distributed Area Scientific Workflows, Alok Singh, Eric Stephan, Malachi Schram and Ilkay Altintas
11:30am Research Paper: Toward Prioritization of Data Flows for Scientific Workflows Using Virtual Software Defined ExchangesAnirban Mandal, Paul Ruth, Ilya Baldin, Rafael Ferreira Da Silva and Ewa Deelman
11:50am Research Paper: On Analytics of File Transfer Rates Over Dedicated Wide-Area Connections, Satyabrata Sen, Nageswara Rao, Qiang Liu, Neena Imam, Rajkumar Kettimuthu and Ian Foster
12:10pm Discussion and Concluding remarks from the chairs
12:30pm Workshop adjourns