Cloudscape
Document Actions
Infrastructure for Service Scalability and High Availability
XtreemOS is working on a number of technologies to help build provide scalable and highly available services for the Grid. Each such technique will be available in the form of a software library. Grid services or applications can then link to these libraries to benefit from the components described in the following pages.
Resource Discovery for Application Execution Management
This research aims at answering an inevitable question when operating a Grid: Where can I find X machines that match criterion Y? This query is necessary as the first step when trying to execute a new job on the Grid.
Instead of relying on traditional hierarchical directory service solutions, we work on a totally decentralized system based on peer-to-peer technologies. This system defines each machine's attributes as a set of coordinates on an N-dimensional Euclidean space. A queries can be represented by a set of coordinates as well, or more likely by an entire sub-spaces (e.g., find 10 machines with 1 to 2GB of memory and a Linux kernel between 2.6.12 and 2.6.18).
Our system somewhat resembles systems like CAN. It however has a number of differences in the way each node is kinked to others. This unique organization is the key to support range queries in a scalable manner.
First software release!
Open position: release manager!
New deliverables released!
