SOSCIP's Advanced Computing Platforms - Call for Proposals
Sponsor
SOSCIP (Smart Computing for Innovation)
Program
SOSCIP's Advanced Computing Platforms [1]
Description
Since 2012, SOSCIP has been Canada’s leading R&D consortium dedicated to harnessing advanced computing to drive industry innovation. Their mission is to bring together industrial partners and academic researchers and provide them with advanced computing technologies and expertise to solve social, technical and business challenges. SOSCIP has built an advanced ecosystem that is integrating state-of-the-art technologies alongside highly qualified personnel (HQP) trained to leverage those technologies and drive Ontario companies to the forefront.
SOSCIP is launching two new HPC platforms to meet the rising demand for GPU-Acceleration and Parallel-CPU access among the SOSCIP community.
Highlights of the two systems are below. For a full description of the platforms, visit the SOSCIP's Advanced Computing Platforms [1] webpage.
GPU-Accelerated Platform
The new SOSCIP GPU-Accelerated Platform is a high-performance GPU cluster powered by NVIDIA Tesla V100 GPUs and IBM Power9 CPUs. Mist [2] is a collaboration between SOSCIP and SciNet, which consists of a total of 54 IBM AC922 servers each with 2×16 core Power9 GPU and 256GB RAM. Each compute node has 4 NVIDIA Tesla V100 GPUs with 32GB of RAM and CUDA capability 7.0 (Volta).This platform supports AI, Machine Learning and Deep Learning frameworks and has the capability of accelerating Molecular Dynamics codes (NAMD, Gromacs, etc.) and other computationally intensive applications in computational chemistry, geophysics, next generation sequencing and astronomy.
Parallel CPU Platform
The new SOSCIP Parallel-CPU platform is a homogeneous high-performance system attached to the SciNet supercomputer Niagara [3], which is designed to enable large parallel jobs in order to optimize throughput of a range of scientific codes running at scale, energy efficiency, and network and storage performance and capacity. Each compute node (based on Lenovo SD530 server) has 40 Intel Skylake/Cascade-Lake cores with 202GB (188 GiB) of RAM. The platform is an ideal system for running parallel code that cannot obtain cost-efficient speedup through GPUs, such as Computational Fluid Dynamics codes (OpenFoam, Nek5000, etc) and Quantum Chemistry codes (VASP, CP2K, etc).
Special Notes
Terms of Use
SOSCIP’s general Terms of Use [4] are available for reference.
Project Collaboration Agreement
SOSCIP projects are governed by a Project Collaboration Agreement (PCA) which describes the responsibilities and obligations of the collaborators and the rights around intellectual property (IP). If a project is approved, SOSCIP will develop the PCA from their standard templates and send it to the academic institutional partner to secure signatures.
SOSCIP Project Access Fee
Once the PCA is fully signed, an invoice for the SOSCIP project access fee in the amount of $5,000 + HST will be sent. The parties to the PCA may wish to share the cost of the SOSCIP project fee.
Note: SOSCIP project access fee may be an eligible expense under NSERC, OCE and Mitacs guidelines.
Deadlines
If College-level review is required, your College will communicate its earlier internal deadlines.
| Type | Date | Notes |
|---|---|---|
Internal Deadline | Please submit your application, along with an OR-5 Form to research.services@uoguelph.ca [5]. The application requires that the Principal Investigator, Co-Investigators, Industry Partners, and Academic Institutions involved in the project all sign the application. | |
External Deadline | Applicant to submit completed application to SOSCIP. |
How to Apply
SOSCIP supports collaborative R&D projects [6] that involve academic and industry researchers working together to solve challenging problems using SOSCIP’s compute platforms.
How to Apply
- Applications must be jointly submitted by a Principle Investigator from any SOSCIP Consortium member institution and their industry partner.
- The SOSCIP team will assist the applicant in early discussions and provide support in partnership building, computing expertise, and alignment with SOSCIP’s mandate.
- Interested applicants should complete the SOSCIP Collaborative Project Application form [7] and share a draft with the Partnerships team. Where appropriate, SOSCIP will help the applicant determine the best funding program (eg NSERC, Mitacs, OCE) for the needs of the project and industry partner.
- Completed applications will be evaluated by the SOSCIP Scientific Advisory Committee [8] based on eligibility, suitability, technology, scientific and commercial merits.
Download the application template [9] and apply for priority access on these new platforms.