1A Computational Economy for Grid Computing

Title: 1A Computational Economy for Grid Computing

Research Question: How can a computational economy be implemented to improve resource management and scheduling in grid computing environments?

Methodology: The researchers developed a resource management system called Nimrod-G for scheduling computations on resources distributed across the world. This system leverages existing technologies like Globus and provides new services essential for constructing industrial-strength Grids. They used an economic paradigm to create an economy-driven grid resource broker that supports soft-deadline and budget-based scheduling of applications on the computational Grid.

Results: They conducted experiments on a world-wide grid testbed that spans five continents. The results showed that the Nimrod-G resource broker was able to dynamically lease Grid services based on users' Quality of Service (QoS) requirements, minimizing time or cost within specific deadline and budget constraints.

Implications: The implementation of a computational economy in grid computing environments can improve resource management and scheduling, making it easier for users to trade off between time for results delivery and economic cost. This can lead to more efficient use of resources and increased system utilization. The use of an economic paradigm can also help create competitive and efficient markets in Grid computing.

Link to Article: https://arxiv.org/abs/0111048v1 Authors: arXiv ID: 0111048v1