Virtual Appliance Marketplace

Average user Rating:254 Customer Ratings (Login to rate)


Plug-and-play solution for building self-configuring virtual clusters delivering high-throughput computing

Features

Collegiate:
Yes

One-line Description:
Plug-and-play solution for building self-configuring virtual clusters delivering high-throughput computing

Filename:
http://www.grid-appliance.org

Size Compressed:
152

Allocated Memory:
256

Username:
griduser

Password:
password

VMware Tools Installed?:
Yes

Operating System:
Debian Linux, kernel 2.6.15.7

Torrent?:
No

Applications:
X11, Condor, IPOP, and gcc, g77, g++ development environment (version 3.3.5).

Last Updated:
September 9, 2007 - 01:53

Description:
Note: this appliance is being enhanced/upgraded frequently to include more features and optimize size. In addition, the documentation has been evolving over time. To download the latest image and documentation, please visit the following site:
http://grid-appliance.org
The Grid Appliance is a self-configuring Virtual Machine appliance that is used to create ad-hoc pools of computer resources over wide area networks to execute high-throughput, long-running jobs. Appliances are connected to each other through a peer-to-peer virtual network using private IP addresses. Upon starting the appliance, it is automatically connected to a pool of resources and is capable of submitting and executing jobs using the Condor Grid scheduler. Its virtual network (IPOP) features decentralized NAT traversal over UDP, a decentralized DHCP service supporting multiple address spaces, and self-configuring Condor pools using a Distributed Hash Table (DHT). Currently, a public infrastructure for bootstrapping such pools is running on PlanetLab; deployments on private resource pools are also supported.
There have been several related projects on similar lines, for instance, SETI@home (http://setiathome.berkeley.edu/), folding@home (http://folding.stanford.edu/), fightaids@home(http://fightaidsathome.scripps.edu/) etc. However, most of these approaches provide only one way job submission capabilities, where the resources of a normal user are just used for some scientific computation without allowing the user any gain in terms of resources. However, our approach fundamentally differs from these in the way that the user gets full access to all the resources of the network and virtually gets the power of a supercomputer on his/her desktop where he/she can submit jobs to the Grid Appliance network.
The sections below describe some of the sample applications and possible usage scenarios for the Grid Appliance.
Development/testing and LAN virtual clusters:
In this scenario, the user can start Grid Appliances on his/her computer network and set one of them as server and others as clients and hence can form a small virtual cluster for himself/herself on his/her personal computer and can use it as a local testbed for development, testing, and self-learning.
The setup can be replicated at a larger extent to form a Virtual Cluster in a LAN environment. Virtual Clusters can be an ideal testbed for the applications and products which are still under the development cycle. Because of the NAT traversal capabilities in the Grid Appliance, it is possible to deploy LAN virtual clusters without the need to assign "bridged" IP addresses to each node - each VM can be configured with a NAT interface.
Virtual Clusters Connected Over WAN
The Grid-Appliance combines virtualization, peer-to-peer routing and NAT/firewall traversal, innovative virtual networking solutions and use of a reliable distributed job scheduler to enable virtual clusters to span across WANs. This setup is useful in collaborations involving multiple institutions or multiple departments within an enterprise.

Technical Specifications

Operating System:

Debian Linux, kernel 2.6.15.7

VMware Tools installed: No

Size: 152MB

Allocated Memory (RAM): 256

Applications Installed:

X11, Condor, IPOP, and gcc, g77, g++ development environment (version 3.3.5).

Virtual Appliance Account Information