ARCHIVED: Getting started on Rockhopper
On this page:
- Introduction
- Requesting an account
- Connecting and logging in
- Managing local and remote files
- Using Modules to set up your software environment
- Compiling programs
- Submitting batch jobs to SGE
Introduction
Rockhopper (rockhopper.uits.iu.edu) is Penguin Computing's
Penguin-On-Demand (POD) supercomputing cloud appliance hosted by
Indiana University.
The Rockhopper POD is a collaborative effort between Penguin Computing, IU, the University of Virginia, the University of California Berkeley, and the University of Michigan to provide supercomputing cloud services in a secure US facility. Researchers at US institutions of higher education and federally-funded research centers can purchase computing time from Penguin Computing, and receive access via high-speed national research networks operated by IU.
For more about On Demand services, see Penguin Computing On Demand. For information about using Penguin On Demand clusters, see Penguin Computing's POD wiki.
Requesting an account
To request an account on Rockhopper, submit the account request form.
Note: You must pay for an account on Rockhopper,
and need a credit card to complete the account request form. To
request an alternate financial arrangement, contact Penguin Computing
directly ( pod@penguincomputing.com ).
Connecting and logging in
Access to rockhopper.uits.iu.edu is via key-based
SSH shell login to head nodes. These keys, along
with the instructions on how to use them, are generated during the
account creation process.
Note: Rockhopper is not an IU resource; you cannot use your IU Network ID to access the POD system.
Penguin Computing also makes available the PODShell remote job submission and data staging tool, which runs on remote Linux servers and personal computers, allowing for remote control of POD jobs.
For POD-specific information about connecting to Rockhopper, see Accessing POD on the POD wiki.
Managing local and remote files
You can move files to and from the POD using SCP and
SFTP. You can also ship detachable hard drives to Penguin
Computing. To make such an arrangement, contact Penguin Computing
directly ( pod@penguincomputing.com ).
Home directories are housed on a local Lustre file system with no quotas.
Note: Rockhopper does not include a separate
scratch file system. Disk usage is a billable item. For more, contact
Penguin Computing directly
( pod@penguincomputing.com ).
Using Modules to set up your software environment
Unlike Big Red and Quarry, the Rockhopper POD uses Modules to control the user's environment.
To see a listing of what modules are available, use:
module availTo load a module, use (replace module name with the
name of the module you want to load):
For POD-specific information about Modules and the user's environment, see Available Applications on the POD wiki.
Compiling programs
GNU, Intel, and Portland Group compilers are installed on the POD. Open MPI compiled with these compilers is available for MPI programs. For POD-specific information about compiling programs see, Compiling Applications. on the POD wiki
Submitting batch jobs to SGE
Users familiar with TORQUE/PBS implementations should
find it easy to work in the Sun Grid Engine (SGE) environment. Often,
qsub parameters are the same between TORQUE and SGE; the
only difference is that #PBS is replaced with
#$. For a complete list of qsub parameters,
see the qsub man page. For POD-specific
information about creating and submitting SGE scripts, see Using
SGE on the POD wiki.
Last modified on May 22, 2013.







