Hi L-A, Cloudman actually uses a shared filesystem as well, there's no local-to-node caching of indices yet. Work is being done to provide more abstract methods for handling galaxy data (see recent ObjectStore commits), though what you're looking for isn't available yet. -Dannon On Dec 9, 2011, at 3:43 AM, Louise-Amélie Schmitt wrote:

Hello,

We're planning on using our lab's cluster for running jobs, but we have to copy genome indices on the local disk where the job is run to avoid too much I/O over the NFS. Since we know Cloudman exists and works fine, we were wondering how the indices are managed in this case, and if there would be a way for us to use a similar solution.

Thanks, L-A ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at:

http://lists.bx.psu.edu/