Transforming Scientific Data Management with Linked Processes

​

In the evolving world of data-intensive research, we aim to rethink how scientific experiments are designed and executed. Our work focuses on creating a self-organizing, linked process ecosystem for scientific data—an alternative to traditional workflow-based methods. By interlinking data and processes, we hope to simplify how experiments are planned and executed, empowering researchers to focus on discovery without being overwhelmed by technical constraints.

​

A New Vision for Scientific Experiments

​

Instead of requiring scientists to design workflows as complex graphs of dependent tasks, we propose an approach that links data through processes at different stages of an experiment. This ecosystem serves as a semantically annotated network, where meaningful data transformations are naturally represented. Workflow management systems remain a part of the ecosystem but act as scheduling tools rather than design frameworks, offering:

• Support for Scientists: Helping researchers identify and schedule the data transformations needed for their experiments.

• Efficient Resource Use: Optimally utilizing computing resources, whether on Grids, Clouds, or other e-infrastructures.

• Scalability for Big Data: Enabling experiments to scale seamlessly as the demands of data exploration grow.

​

Areas of Research

​

Our work touches on several areas critical to advancing this vision:

​

1. Policy Enforcement for Secure Data Exchange: Ensuring data sharing complies with security and privacy regulations.

2. Advancing Multimodal Synthetic Data Generation: Generating synthetic datasets to support scalable, privacy-aware research.

3. Computing in the Web Browser: Developing methods to run computations directly in web browsers for greater accessibility.

4. Beyond Scientific Workflows: Networked Open Processes: Exploring more flexible, open alternatives to traditional workflows.

5. Large Object Cloud Data Storage Federation: Enhancing the integration and accessibility of large datasets across cloud systems.

6. SDN-Aware Data Transfer for Scientific Applications: Using Software-Defined Networking to optimize data movement for demanding applications.

​

Building a Collaborative Future

​

This research is not just about solving technical challenges; it’s about enabling collaboration, discovery, and innovation in a way that feels natural for scientists. By connecting data and processes meaningfully, we hope to create tools and systems that make research smoother and more productive.

​

For more information about this work and related projects,