Established as a collaboration between Carnegie Mellon University (CMU) and King Mongkut's Institute of Technology Ladkrabang (KMITL), CMKL University will integrate two DDN AI400X, the all-NVME flash appliances within the A³I product line of AI and analytics data solutions, with 645TB of raw flash capacity and 3,2PB of raw disk capacity to ensure maximum productivity of the six NVIDIA DGX A100 systems. CMKL selected DDN because of performance, multi-rail network, the small footprint, and the trusted solution provided by DDN and NVIDIA demonstrated through the NVIDIA DGX Reference Architecture. New test results using NVIDIA A100 GPUs for deep learning report that higher throughput can directly translate to higher application performance and shorter run times, which allows institutions like CMKL to get 50 percent improved performance from their AI infrastructure compared to traditional NAS storage.
"Undeniably, since AI has entered our lives, it has greatly facilitated our social development and research capabilities, and has facilitated breakthroughs across all industries", stated Dr. Akkarit Sangpetch, ECE Programme Director at CMKL University and a member of KMITL's Faculty of Engineering. "AI, furthermore, has proved its great ability to identify patterns in the data, and we are committed to offering the fastest and most sophisticated infrastructure to unlock insights contained in those massive datasets. DDN's solution gives us the ability to do just that. In addition, their trusted reputation and deep experience with educational institutes further affirm our goal in providing an AI data platform for research and education."
"As a part of the university-based research infrastructure to drive Thailand's competitiveness in digital and AI technology, we aim to stimulate practical research and development projects that will help jumpstart our AI economy", stated Dr. Wannarat Suntiamorntut, Chair of Digital Research Subcommittee, Competitive Programme Management Unit, which supports the project as an initiative by the Ministry of Higher Education, Science, Research and Innovation.
DDN A3I solutions are fully optimized to achieve the most from at-scale AI, analytics and HPC applications running on NVIDIA DGX systems. They are designed to provide extreme amounts of performance, capacity and capability through a tight integration between DDN and NVIDIA platforms, including the DGX A100. Every layer of hardware and software engaged in delivering and storing data is optimized for fast, responsive and reliable access. This unmatched performance will propel CMKL research and allow students in autonomous driving, precision health, computational flow dynamics, physics and other data-intensive fields, with faster access to ever larger data sets, which is crucial to meaningful research.
"NVIDIA DGX systems simplify deployment and streamline AI development workflows, creating an end-to-end infrastructure solution that can be operational in weeks instead of months", stated Raymond Teh, Vice President, NVIDIA Asia-Pacific. "DDN A3I with NVIDIA DGX A100 provides the power and speed required to advance the important work of the AI researchers at CMKL University."
The DDN A3I with NVIDIA DGX A100 solution helps academic institutions build systemized leadership-class infrastructure with the high-performance storage needed to drive academic discovery based on data analysis and complements existing simulation-based methods, increasing precision of results.
"We are proud to serve as the dependable storage infrastructure provider for CMKL University", stated James Coomer, senior vice president of products, DDN. "With DDN cutting-edge storage solutions, CMKL is fully leveraging at-scale data and turning it into a strategic asset supporting scientific breakthroughs that can drive increased funding, enhance recruiting and improve the lives of countless people around the world."