Abstract:

The term "the Grid" was coined in the mid 1990s to denote a proposed 
distributed computing infrastructure for advanced science and 
engineering.  Less than five years later, it feels as if "the Grid" 
is everywhere -- grid related research and development projects are 
popping-up all over the globe.  They come in different sizes and 
scopes: some are small, others are international; some focus on pure 
computer science problems, while others are interdisciplinary.  
Most of the middleware developed by the different research projects is 
released to the community under an "open source" public license. 
We will present our perspective of this recent wave of Grid-driven 
activities and middleware and discuss how they relate to our ongoing 
work in the area of High Throughput distributed computing. 
The talk is based on our decade-long experience with the Condor high 
throughput computing system that is widely used in academia and 
industry (www.cs.wisc.edu/Condor). We will discuss the challenges 
clients of Grid resources face in harnessing the power of these 
resources and outline the design principals and capabilities of our 
client tools.  Our tools are designed to meet the needs of large 
Master Worker applications and to leverage the capabilities of the 
Condor-G job manager.