Traditional electronic storage solutions for Rice faculty, staff, and students proved insufficient for massive data sets amassed by researchers starting around 2010, and several Rice researchers turned to external solutions. After free solutions were exhausted, researchers began returning to Rice for storage solutions. As BCB Faculty Fellow Dr. Daniel Harrington explained, “Everybody’s Dropbox is getting full and we don’t want to have to pay for more storage.” The results of an IT Task Force survey also indicated that university-funded storage for research data was a high priority for Rice faculty.

In 2013, IT began exploring storage options and ran a trial program in the fall. Rice contracted with Box.com, an external collaboration and storage service provider, to offer 30 gigabytes (GB) of electronic storage to all Rice faculty and staff. Rice Box rolled out shortly after spring break 2014 as the first of several new storage solutions for faculty and staff.  Rice undergraduates and graduate students will be able to share files with their faculty sponsors in the fall.

Rice Box is a U.S.-based cloud solution. This means that although the data saved in Rice Box is not stored in the university’s footprint, it is stored somewhere in the United States. This is important from a security aspect. External service providers often offer free storage solutions, but the data may be housed anywhere around the world, and their server clusters transfer files from base to base as storage needs fluctuate. For sensitive and confidential information like grant proposals and collected research data, the uncertainty of where the data lives is unacceptable.

The most important feature that Rice Box offers participants is collaboration. Documents can be shared with Rice community members as well as external partners. When a document is updated, all participants can receive a notification by email. Comments are also shared and saved inside Rice Box, so it is easy to login and see the most recent activity. “The standard way of collaborating on a grant proposal was by emailing updates to all the participants. So you’d get version 7, no wait, was that 7a or 7b?” said Harrington. “It was very confusing. So with Box, you login and here’s the most current version. We store all the former versions in a ‘Previous Versions’ folder. A lot of our research collaborators are with other universities, so this is much better than having one administrator and wondering what version [of a document they are using]. The transparency and access are great.” For Rice researchers like Harrington, who has used cloud-based collaboration for more than 20 grant proposals and research papers, tracking only the most current version of data and documents is critical.

“And we’re also using OWL-Space for research projects,” he said. “Once we found out that you can use OWL-Space for collaborative projects, and that it comes with a gigabyte of space per project, we began requesting logins for our external collaborators. It became a great additional solution that replicated much of Dropbox’s functionality.”
For everyday work on projects, “Box has been a great way for me to store and access my instrument procedures,” added Harrington. “At Rice, users of the SEA (Shared Equipment Authority) are required to bring their own documentation when they use the instruments. But in the Clean Room in Abercrombie, there is a ‘no paper’ rule due to particulate release from the paper fibers. So I pull out my iPhone or iPad and look up the manual I’ve saved in my Box account.”

Successful grant submissions and partnerships with the Ken Kennedy Institute for Information Technology (K2I) and education-friendly corporations like IBM have allowed the Research Computing Support Group (RCSG) to offer multiple platforms and clusters on which Rice research teams can build and run their serial and parallel computing jobs. However, there were three main obstacles to overcome in 2013:

• Evaluate aging clusters for replacement

• Train graduate students about RCSG resources

• Find storage solutions for active use and archiving

Retiring and replacing aging supercomputer clusters was perhaps the most straightforward challenge. The oldest cluster is now scheduled for retirement in April 2014.

With faculty encouragement and support from the IT Task Force survey, RCSG began offering 30-60 minute workshops once or twice each month.  Primarily directed at graduate students, the workshops are open to anyone interested in the session topic which ranges from Anatomy of a Cluster Job to Advanced Secure Shell for Power Users. Chandler Wilkerson, High Performance  Computing (HPC) Cluster Architect, organizes the workshops and leads most sessions, and he also invites topic suggestions from the audience.

For storage solutions, several were considered and some were tested with pilot participants.  SPICE was the obvious internal storage solution for researchers for big data research projects, but a revision of the charging model was required to meet grant funding requirements (see sidebar). Data management, another aspect of federal grant funding requirements, had been satisfied in 2012 through a collaboration of Fondren Library and RCSG, but many Principal Investigators (PIs) were still unaware of the new Data Management resources.  A cloud collaboration solution, Box.com, was the most likely solution for active collaboration for both faculty and staff and included features for easily sharing documents with external collaborators.  This solution was tested with a pilot group in fall 2013. Rice contracted with Box to supply U.S.-based Internet2 NetPlus Service for Rice faculty and staff in 2014.

Working with Rice’s Research and Cost Accounting (RCA), IT redesigned the SPICE (Shared Pool of Integrated Computing Environments) charging model to fit requirements established by grant funding organizations such as the National Institutes of Health (NIH) and the National Science Foundation (NSF).  Although Rice researchers receive a designated amount of storage space for their projects, researchers in fields requiring big data and high performance computing clusters often require additional space, or they may need a virtual machine (VM) in which to run their code.  The new charge model approved by RCA included storage increments called bricks.  A VM Brick consists of either one, two or 4 cores and 40 Gigabytes (GB) of disc space, reserved for the operating system.  A Storage Brick consists of one Terabyte (TB) of storage.

The storage space and VMs purchased under the new SPICE model are applied to solutions already bought by researchers, so there are no additional charges, only a revised charging model. SPICE appeases the researchers who need any number of TBs of space (minimum of one) and/or a research computing space separate from their desktop computer, but do not need the more vast resources for High Performance Computing (HPC) provided through the Research Computing Support Group (RCSG). However, it should be noted that the VMs in this model are meant to handle moderate to low computations.

“SPICE, a Shared Pool of Integrated Computing Environments, represents an evolutionary step in providing centrally-managed storage and servers to Rice researchers. It represents a model to achieve our “condo computing” model into the server and workstation based research arena. SPICE provides to Rice University’s Researchers a cost and time saving ability.” said HPC Support Supervisor, Joseph Ghobrial.

On July 28, 2014, we migrated the STIC scheduler/resource manager from MOAB/TORQUE to the open-source  solution SLURM (Simple Linux Utility for Resource Management (http://slurm.schedmd.com/)).

While MAUI/TORQUE works well for uncomplicated queuing requirements like those of SUGAR and BlueBioU, SLURM is required for more complicated condo environments like STIC and DAVinCI.

SLURM was developed by a group at the Lawrence Livermore Nation Laboratory(LLNL) and is now required in all large scale lab procurements which forces system vendors to get behind SLURM in order to bid on national lab procurements.  Many of our academic colleagues (Harvard, Duke, Oxford, Michigan, Buffalo, etc.) have already moved to SLURM.

For more information about SPICE, visit: http://www.rcsg.rice.edu/spice/

For more information about the SLURM migration, visit: http://www.rcsg.rice.edu/stic-migrating-from-torquemoab-to-slurm/

In Rice’s Chevron Visualization Laboratory viewers can see everything from molecules to geological images and even interact with the data to look for correlations on the NSF-funded DAVinCI visualization wall. The lab features a 200-inch wall that allows for displays of data sets for all disciplines, and is capable of displaying over 32 megapixels. Special glasses can make the images three dimensional. The wall is constructed from 16 high-resolution projection monitors and custom graphics engines and is powered by DAVinCI (Data Analysis and Visualization Cyberinfrastructure), one of Rice’s high performance clusters, consisting of 2,304 processor cores in 192 nodes.

Erik Engquist, Visualization Project Manager, works with faculty and  Houston area researchers to prepare the data and interface with the technology so they can focus on the research.  Office hours and demonstrations are listed online (viz.rice.edu).