Version 3 by Daniel Pop
on Mar 31, 2014 23:38.

compared with
Current by Daniel Pop
on Mar 31, 2014 23:46.

Key
This line was removed.
This word was removed. This word was added.
This line was added.

Changes (3)

View Page History
* a Monitoring/Quality measuring tools: the integrated Puppet Configuration Management system natively provides capabilities for integration with and deployment of the [Nagios|http://www.nagios.org]monitoring solution. This allows operators/administrators to provide a better QoS.

Next section presents SCAPE Cloud Deployment Toolkit, as name suggests, a set of tools that supports automatic deployment of SCAPE components on multiple Cloud environments, thus enabling portability of SCAPE software between clouds.

h5. Cloud Deployment Toolkit for SCAPE Platform

The Cloud Deployment Toolkit provides system administrators with a user-friendly GUI that uses an orchestration layer for deploying common SCAPE components and tools on virtual machines hosted on Eucalyptus or Amazon EC2 (for the moment these are the environments we tested, but we can extend this to other clouds supported by the orchestration layer). The role of the orchestration layer (that is [Apache Libcloud|https://libcloud.apache.org/]) is to hide the differences between the above mentioned Cloud vendors (providers) such that these differences are nonexistent from the user's points of view and moreover to minimize or even eliminate the vendor lock-in. A major advantage from using multiple providers would be the increase in availability for the SCAPE platform and also the possibility of reducing costs by testing different scenarios on Eucalyptus, and switch to usage of Amazon EC2 only when necessary (e.g. if Eucalyptus is down or has performance problems).

Together with Powered by Puppet and Puppet DB, the deployment of common SCAPE components and tools leaves human interaction out, such that the deployment will be the same on all nodes (which use the same Puppet templates) regardless of the selected Cloud vendor, thus reducing the overhead produced by debugging and testing the result of the deployment on different nodes. Concretely, the toolkit provides features that allows:
* auto-detecting the Eucalyptus environment,
* listing existing clusters,