Abstract
Grid is a collection of distributed resources available over a local or wide area network that appear to an end user or application as a large virtual system. The implementation of individual applications on a grid promises performance benefits over that achievable at a single site for many resource-intensive applications. However, with the emergency of OGSA (open grid services architecture), scheduling issues face new challenges in the new grid environment. Although higher-level abstraction makes resources encapsulated in a uniform fashion, it also leads to slower response in the process of execution. A new scheduling mechanism is needed to exploit characteristics of OGSA. In this paper, we put forward an effective instance-based application scheduling mechanism, implemented in Globus Toolkit 3. Good performance, best resource management and good lifecycle control in this mechanism can also be guaranteed.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
