The Office of Advanced Research Computing now offers Buy-In to Rutgers’ next generation of high-performance computing

We are excited to announce the next phase in Rutgers Advanced Computing. During the Spring Semester of 2017 (March or April) the Office of Advanced Research Computing (OARC) is unveiling Amarel, a “condominium” style computing environment developed to serve the university’s wide-ranging research needs.  The Amarel Cluster provides a shared platform that optimizes resources for the benefit of all users.  Named in honor of Dr. Saul Amarel, one of the founders of the Rutgers Computer Science Department and contributor to advanced computing and artificial intelligence research and methodologies, is designed to suit many different research applications.

Amarel is a shared community-owned advanced computing environment available to any investigator or student with projects requiring research computing resources.  The model is distinguished by a set of uniform technical and policy elements that, while sufficiently flexible to support the degree of heterogeneity inherent in the Rutgers environment, affords a level of standardization that greatly simplifies management procedures, troubleshooting, and provides a more seamless user experience.

Investigators can become condominium owners through a University-sponsored matching program.  Owners have highest priority and are guaranteed access commensurate with their investment. Owner allocations are designed to feel as close to a locally maintained and managed system as possible, but with unused cycles being available to the general pool.  Free access is available on a first come first served basis to the Rutgers community utilizing general pooled resources under an open access policy. Non-owner jobs can be preempted immediately by higher priority jobs.

OARC is currently looking for researchers, departments, and schools interested in participating in the first buy-in of the Amarel system. More information on node configurations and estimated costs can be found on the OARC website, oarc.rutgers.edu/services/condo-model.

With OARC covering costs of racks, cabling, network interconnect, system administration, maintenance, and spare nodes, we offer very competitive pricing. Because money is used for purchasing equipment, indirect costs do not apply.

The OARC support team comprises six highly experienced cluster system administrators and four PhD-level computational scientists with deep expertise in the STEM and life sciences.  It is this unique combination of technical capability and scientific support working collaboratively with Rutgers investigators that enables problems to transition toward solutions at an accelerated pace.  Support and collaboration includes proposal development, post award participation, training, data management services and both causal and formal consulting services.

OARC anticipates that the greatest growth will occur within the condominium cluster as it will demonstrate the sustainability qualities that are essential for Rutgers to continue to provide its investigators with state-of-the-art resources for computationally demanding problems. While the requirement for specialized project or discipline oriented equipment will continue to exist, the appeal of University-subsidized computational resources supported by expert technical and computational science personnel has and will prove to be a great incentive to make community investments. The Amarel cluster is designed to meet the needs of a broad and diverse user community by being flexible enough to support

  • Traditional High Performance Computing
  • High Throughput Computing (HTC)
  • Data Analytics
  • Large memory systems
  • Virtualization
  • Accelerators (GPGPUs, XEON PHIs, FPGAs)
  • The ability to couple with edge devices

Research computing at Rutgers will also embrace connections to facilities on national research networks, cloud-based resources such as Amazon Web Services, Google Cloud Platform, and Microsoft Azure, and cooperative facilities such as Open Science Grid, and additional resources shared among our neighboring research institutions.  As OARC collects data on cost, reliability, performance, and overall user experience of local and remote access options, we expect to further optimize the way we map computing requirements to available resources.  The ultimate objective is to consistently provide the most efficient, cost-effective and supported solutions to the research community to move from problem statement to solution as expeditiously as possible.

All OARC technical facilities are maintained using recognized security principles, practices and tools.  To the extent possible, OARC managed system adhere to the same design and configuration standards, thus simplifying operational procedures, updates and general security.

 

If you have questions regarding Amarel, contact us:                                                                                                                                                                                                                                            Kristin Lepping, Assistant Director ♦ (848) 445-5140 ♦ klepping@rutgers.edu