We are glad to inform you that, GUJCOST has established the Supercomputer facility at the LDRP Institute of Technology and Research, Gandhinagar for providing the platform to the young students, research scholars and academic researcher with innovative ideas and to provide an end-to-end solution for high-performance computing (HPC) / deep learning, starting from latest hardware, GPU Accelerated software development environment, application support, training and tutorials to nurture and satisfy needs of the end users. The facility aims to provide computational resource with advanced technologies to perform high-end computations for scientific, engineering and academic programmes to address and catalyze the research using modeling, simulation and data analysis.
The evolving face of technology has necessitated a fresh relook at our technology policies and investments. The world itself is witnessing a new revolution brought on by technology and it is continuing to change our lives in every way. Today, supercomputing has become a new reality in the new world order. In fact, the growth potential of a nation is measured by the availability of its supercomputing resources.
Supercomputing today has become the new paradigm for growth and development all across the world. It is the key to new reforms in all societal sectors of the economy from manufacturing, defence, power, agriculture, communications, banking, and so on.
Objectives of the Supercomputer Facility :
- Create a pool of skilled professionals for pursuing research and development in the area of HPC and promote the use of HPC systems and parallel programming in several application domain areas.
- Creating employment opportunity students / researchers for R&D organization and industry.
- Disseminating advanced computing techniques useful in many research academic semester projects and Ph.D. thesis.
- Introduction of HPC as a subject in the academic/engineering the curriculum at UG/PG level.
- Providing a base of Supercomputing expertise to develop interdisciplinary computational science and deep learning degree programs.
- Help professor and faculty members in developing funding proposals for HPC research-based projects, providing not only a set of resources to list in proposals but also expertise on emerging issues that are of interest to funding agencies.
- Encouraging possibilities of various national and international labs for joint research in HPC arena.
Programs and Activities :
- Faculty and students across all academic/engineering disciplines can be educated in computational science and deep learning techniques, and thereby positioning them for careers in cutting edge science and technology.
- Faculty and Researchers who employ HPC in their engineering software can be educated in software design methodologies that allow maximal exploitation of computing resources for complex problem-solving environments.
- Faculty and students can incorporate into their research codes the techniques of sound software design, coding, debugging, testing, maintenance, porting and performance evaluation.
- Students can be taught to act as consultants to application teams that use supercomputing and deep learning techniques by providing a pathway for disseminating knowledge.
- Student researchers can have opportunities to present ongoing projects to an audience of HPC educated colleagues, in order to obtain feedback on improving their investigations.
- Interactions with relevant stakeholders including industry, academia and students from other colleges and universities.
- Faculty and researchers can apply for and obtain funding for project proposals involving supercomputing researches.
PARAM Shavak – Supercomputing Solution in a Box
Param Shavak machine PARAM Shavak – solution, aims to provide computational resource (Capacity building) with advanced technologies to perform high-end computations for scientific, engineering and academic programs to address and catalyze the research using modelling, simulation and data analysis. This initiative is expected to create HPC aware skilled workforce (Capability building) and for promoting research by integrating leading-edge emerging technologies at grass root level. As the scope and complexity of computational needs continue to increase at colleges universities, professors and administrators are compelled to seek appropriate and affordable solutions. PARAM Shavak provides the computing power necessary to keep academic institutions on the leading edge in today’s competitive market at affordable cost. This system is meant for research organizations and academic institutions that are on the verge of adopting HPC culture in their institutions/organizations. Besides a handful of value additions from C-DAC, the system comes with most of the features that can be found in a full blown HPC clusters job schedulers, compilers, parallel libraries, mpi, resource managers, some of the commonly used HPC applications in engineering scientific domains etc.
“Watch The PARAM Shavak Video”
(File Format: MP4, Size: 12.6 MB, Date: 17/03/2016, Playing Time: 01.58)
PARAM Shavak Architecture
The system consists of 2 multicore CPUs each with minimum 12 cores along with two number of accelerator cards. The entire configuration is available in a single server in a table top model. Regardless to the traditional HPC systems/supercomputers, this system does not require specific support infrastructure like precision air-conditioned environment, controlled humidity etc. Also, the accepted sound level is very less when compared to the traditional servers. This brings down the infrastructure cost of the system.
Salient Features
- Supercomputer in a Box system in a table top model
- Powered with 2 multicore CPUs each with minimum 12 cores
- 2 numbers of accelerator cards
- 3 Tera-Flops peak computing power with 8 TB of storage
- Easy to deploy solution with no additional datacenter infrastructure
- Pre-loaded with parallel programming development tools and libraries
- Preinstalled tools for scheduling and resource management
- Video Tutorials, learning materials and user manuals etc.
- Indigenous HPC portal- CHReME
- Indigenous applications interface – Onama (File Format: .pdf, File Size: 4.23 MB, Date: 24/02/2016)
Built-in HPC Applications
- Bio-informatics: mpiBLAST
- Molecular Dynamics: GROMACS
- Materials Science: Quantum Espresso
- Quantum Chemistry: NWChem, ABINIT
- Atmospheric and Ocean Modeling: WRF, MOM
- CFD: Open FOAM