« Back
Bridges to the Future
Bridges to the Future
Bridges, available starting in 2016 at the Pittsburgh Supercomputing Center, is a new concept in high-performance computing—a system designed to support familiar, convenient software and environments for both traditional and non-traditional HPC users. It is a richly-connected set of interacting systems offering a flexible mix of gateways (Web portals), workflows, Hadoop and Spark ecosystems, interactivity, and batch processing. Bridges will include:
- 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 diverse applications, for example, in machine learning, image processing, and materials science
- Database nodes to support data management, analytics, integration, and fusion and to drive gateways and workflows
- 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
Bridges could be a good fit for you if:
- You want to scale your research beyond the limits of your laptop, using familiar software and user environments.
- You want to collaborate with other researchers with complementary expertise.
- Your research can take advantage of any of the following:
- Rich data collections: Rapid access to data collections will support their use by individuals, collaborations and communities.
- MPI jobs requiring up to 1176 cores (ensemble and other loosely-coupled jobs requiring more cores in aggregate can exceed 1176 cores).
- Cross-domain analyses: Concurrent access to datasets from different sources, along with tools for their integration and fusion, will enable new kinds of questions.
- Gateways and workflows: Web portals will provide intuitive access to complex workflows that run "behind the scenes."
- Large coherent memory: Bridges' 3TB and 12TB nodes will be ideal for memory-intensive applications, such as genomics and machine learning.
- In-memory databases: Bridges' large-memory nodes will be valuable for in-memory databases, which are important due to their performance advantages.
- Graph analytics: Bridges' hardware-enabled shared-memory nodes will execute algorithms for large, nonpartitionable graphs and complex data very efficiently.
- Optimization and parameter sweeps: Bridges is designed to run large numbers of small to moderate jobs extremely well, making it ideal for large-scale optimization problems.
- Rich software environments: Robust collections of applications and tools, for example in statistics, machine learning and natural language processing, will allow researchers to focus on analysis rather than coding.
- Data-intensive workflows: Bridges' file systems and high bandwidth will provide strong support for applications that are typically I/O bandwidth-bound. One example is an analysis that runs best with steps expressed in different programming models, such as data cleaning and summarization with Hadoop-based tools, followed by graph algorithms that run more efficiently with shared memory.
- Contemporary applications: Applications written in Java, Python, R, MATLAB, SQL, C++, C, Fortran, MPI, OpenMP, OpenACC, CUDA, and other popular languages will run naturally on Bridges.