Compute Through Bridges

Bridges is a high performance computing system designed to support familiar, convenient software and environments for both traditional and non-traditional HPC users. Its richly-connected set of interacting systems offers exceptional flexibility for data analytics, simulation, workflows and gateways, leveraging interactivity, parallel computing, Spark and Hadoop.

All Brain Image Library data is directly accessable on bridges.psc.edu and is located on the /bil file system.

Datasets in BIL are uniquely referred to by a combination of a 16 digit unique identifier and the dataset name separated by a period: <16-digit-unique-identifier>.<Dataset-Name> The file path to BIL entries will be: /bil/data/<c1c2>/<c3c4>/<16-digit-unique-identifier>/<Dataset-Name>

where <c1c2> are replaced by the first and second characters of the unique identifier and <c3c4> are replaced by the third and fourth characters of the unique identifier. For example, the dataset: abcded0123456789.example_dataset_01 would be found down the following path: /bil/data/ab/cd/abcdef0123456789/example_dataset_01/

Additional information about Bridges can be found here.

Get access to Bridges:

You can get access to Bridges in two ways: apply for an allocation if you are faculty, a researcher or educator with a US institution, or be added to an existing account if you work with a PI who has an existing Bridges allocation.

---

Apply for an allocation

We offer two programs to provide access to Bridges. Complete allocation and affiliation information can be found on the allocations page, but in a nutshell:

• Researchers and educators at a US academic or non-profit research institutions can apply for Bridges access through the NSF's XSEDE program. The Bridges FAQ provides more information on how to apply to Bridges through XSEDE. Three types of access are available:
  • Start-up allocations allow you to explore high-performance computing and Bridges. The application process is streamlined and you can request up to 2500 GPU-hours on Bridges GPU nodes, up to 50,000 core-hours on Bridges' "regular" (128GB memory) nodes, up to 1000 TB-hours on Bridges "large" (3 and 12TB memory) nodes, or any combination of those resources. See https://portal.xsede.org/allocations/startup for details on applying.
  • Research allocations provide expanded resource limits when you are ready to scale up your work. See https://portal.xsede.org/allocations/research for details on applying.
  • Coursework allocations supplement educational activities from workshops to semester courses that benefit from the use of high-performance computing. See https://portal.xsede.org/allocations/education for details on applying.

• Proprietary research can be accommodated through the Corporate Affiliates program.

 

Get added to an existing allocation

You can be added to an existing allocation by the PI. You must have an XSEDE Portal account (create one here if you don't already have one); once you do, the PI can submit a request through the Portal to add you to the allocation. The request must specify all resources which you should be given access to.

Every Bridges user needs access to Pylon storage, but it is not granted automatically; it must be specifically requested in the Add User process.

In addition, the PI must specify any other Bridges' resources granted to the allocation that you should have access to. These can include RM nodes, GPU nodes and/or LSM nodes.

Choose your tools

Bridges is a uniquely capable resource for empowering new research communities and bringing together HPC, AI and Big Data. It is designed to support familiar, convenient software and environments for both traditional and non-traditional HPC users. Its richly-connected set of interacting systems offers exceptional flexibility for data analytics, simulation, workflows and gateways, leveraging interactivity, parallel computing, Spark and Hadoop.

A sampling of Bridges' tools which will enable your work includes:

• Compute nodes with hardware-supported shared memory ranging from 128GB to 12TB per node to support genomics, machine learning, graph analytics and other fields where partitioning data is impractical
• GPU nodes to accelerate applications as diverse as machine learning, image processing and materials science
• Database nodes to drive gateways and workflows and to support fusion, analytics, integration and data management
• Webserver nodes to host gateways and provide access to community datasets
• Data transfer nodes with 10 GigE connections to enable data movement between Bridges and XSEDE, campuses, instruments and other advanced cyberinfrastructure

 

Get more information

If you would like to know more about Bridges and how you can use this unique resource in your research, see the "Applying for a Bridges account" FAQ.

If the FAQ doesn't answer your questions, call us at 412-268-4960 or email bridges@psc.edu.