Small Business Innovation Research/Small Business Tech Transfer
Accelerating HPC Applications through Specialized Linear Algebra Clouds
Project Description
Cloud computing has the potential to permit scientists to scale up to solve large science problems without having to invest in hardware and software infrastructure. Even though its use has become commonplace in the business realm, its use in HPC is still limited to a few applications. This project aims at developing, prototyping, validating, and commercializing the world's first cloud-enabled acceleration library of linear algebra solvers. Considering that linear algebra is the most common computational bottleneck in scientific software applications (more than 70% of HPC cycles), the resulting Specialized Linear Algebra Cloud technology will directly impact numerous applications, such as the design of aerospace and automotive vehicles, exploration for new oil & gas reserves, nuclear energy research, and the design of computer chips, among many others. The Phase I work will incorporate innovations for removing the bandwidth bottleneck in the outsource model of a Specialized Linear Algebra Cloud. We will also design a core platform able to support solver-as-a-service technology in HPC environments. The Phase I technology will be evaluated in three different scenarios: (A) Infrastructure Cloud outsourcing computations to Specialized Clouds, (B) Infrastructure Cloud making use of internal Specialized Clouds, and (C) End-users accelerating code through direct access to Specialized Clouds. Accelogic's novel numerical libraries are expected to provide next-generation speed/efficiency to existing cloud systems, with direct application to NASA programs, other government applications, and many valuable uses in academia and the private sector. The Phase I proof-of-concept work should advance the technology from TRL 2 to TRL 3, and we expect to reach TRL 5 during Phase II. A world-class combination of experts in algorithm design, computer networks, and numerical analysis pursue these development goals.
More »
Anticipated Benefits
For other government customers (e.g., DOD) and private-sector users, the new technology will be integrated into Accelogic's HPC product line, LAPACKrc, which is Accelogic's heterogeneous library (currently under development) for solving linear algebra problems. This SBIR addresses the Internet/Intranet-based linear equation solver server that will allow users to submit jobs remotely and to pay on a job basis with no new hardware costs. Large companies and government/academic/corporate supercomputer centers will purchase the server technology and serve its own users. Small and medium-sized companies needing "speed on a limited budget" will benefit from purchasing service from the availability of such servers on a job basis. Technology-driven companies that must solve large-scale linear equation problems, such as those in structural design and computational chemistry, along with the Computer Aided Engineering (CAE) industry, supercomputer manufacturers and industries providing cloud services will also generate demand.
NASA's supercomputer users must solve large-scale linear algebra problems faster and cheaperyet with guaranteed accuracy. The linear-algebra acceleration cloud we envision producing during this SBIR project will meet this NASA need in the near term. We expect to generate a value proposition that results in a Phase III NASA acquisition (with cloud computing centers procuring our software, mainly through our hardware partners, as an add-on to their hardware enhancement purchases.) Accelogic could then sell an integrated solution that has a clear procurement channel, and that will work on Day 1. (Our Phase III partners will work closely with us to ensure seamless/high-performance integration.) In particular, the framework described in the proposal will enable NASA to take full advantage of cloud-enabled petascale computing by providing a set of tools that can be readily integrated into current (and future) applications by replacing standard calls to state-of-the-art solvers like ScaLAPACK and PETSc. More »
NASA's supercomputer users must solve large-scale linear algebra problems faster and cheaperyet with guaranteed accuracy. The linear-algebra acceleration cloud we envision producing during this SBIR project will meet this NASA need in the near term. We expect to generate a value proposition that results in a Phase III NASA acquisition (with cloud computing centers procuring our software, mainly through our hardware partners, as an add-on to their hardware enhancement purchases.) Accelogic could then sell an integrated solution that has a clear procurement channel, and that will work on Day 1. (Our Phase III partners will work closely with us to ensure seamless/high-performance integration.) In particular, the framework described in the proposal will enable NASA to take full advantage of cloud-enabled petascale computing by providing a set of tools that can be readily integrated into current (and future) applications by replacing standard calls to state-of-the-art solvers like ScaLAPACK and PETSc. More »
Primary U.S. Work Locations and Key Partners
| Organizations Performing Work | Role | Type | Location |
|---|---|---|---|
| Accelogic, LLC | Lead Organization | Industry | Weston, Florida |
Ames Research Center
(ARC)
|
Supporting Organization | NASA Center | Moffett Field, California |
Primary U.S. Work Locations
-
California
-
Florida
Suggest an Edit
Recommend changes and additions to this project record.
This is a historic project that was completed before the creation of TechPort on October 1, 2012. Available data has been included. This record may contain less data than currently active projects.