Java in the Cloud

Cloud computing has a problem. Well, it's not so much a problem; it's more of a question of perception.

Potential cloud converts appear to be put off by concerns over security inside multi-tenant clouds, compliance concerns on where cloud data physically resides and all manner of misgivings relating to how easy applications will be to administer and manage once they have been hosted in a serviced-based environment.

In something of an attempt to allay some of the fears that do exist, several members of the OpenStack community have announced their collaboration to build TryStack, a free sandbox for those interested in exploring and testing the open source cloud operating system.

Initially driven by OpenStack members Rackspace, Dell, HP and also NTT, TryStack is designed for developers and users interested in testing software that communicates over OpenStack APIs without having to set up and administer their own deployment.

TryStack users will be able to launch OpenStack Compute instances that last up to 24 hours, at which point resources for the instances are reclaimed and made available to other TryStack users. Each registered user will receive a set amount of Stack Dollars they can use to "lease" instances within that time period.

"As an OpenStack contributor, I am happy to see TryStack available for those looking for a starting point to explore OpenStack's ecosystem of distributions and products," said Jesse Andrews, director of development for Rackspace Cloud Builders. "TryStack provides a simple way to get some hands-on experience with OpenStack."

TryStack is intended to provide users with the ability to launch instances in one of several TryStack zones, representing different OpenStack reference architectures and geographical locations.

The project also serves as a place for the OpenStack development community to proactively identify any problems with packaging and deployment. Additionally, end users can gain experience administering an OpenStack cloud on a variety of heterogeneous hypervisors and network topologies, as well as document differences in behaviour, functionality and performance between various reference architectures.