QCT will be participating in NVIDIA’s GTC 2022 online event taking place from March 21-24 to discuss hot topics spanning AI, HPC, data center, and cloud computing with speaking sessions presented by our experts.
With the exponential growth of data comes the need to process, store, and analyze that data which requires many industries to revise their infrastructure and environments. Building an integrated workload-driven system with optimized performance can be time consuming while demanding on many resources whether it’s for AI or HPC workloads. At GTC 2022, QCT has put together a digital tour to showcase out latest solutions and products to address these pain points.
Traditionally, applications have relied on numerical methods for computing on high performance computing (HPC) infrastructures; today we see more users running GPU-accelerated computing on HPC infrastructures. This gives rise to heterogeneous architectures which are challenging to accelerate performance more effectively. QCT’s Platform on Demand (QCT POD) are rack level systems with best practice hardware and software integration for specific workloads designed for High Performance Computing (HPC), Artificial Intelligence (AI) and Data Analytics. Utilizing common building blocks, they have greater flexibility and scalability to meet business demands from different industries.
[S42272] QCT POD Tackling AI/HPC Challenges and Boost System Performance
Stephen Wheat, Senior Director, HPC, CTQ
This session will elaborate how QCT built up an adaptive workload driven QCT POD solution with a pre-compiled workload package, rapid deployment, and easy management features. To help different industrial users to prepare their HPC/AI systems effectively and accelerate their system performance. We’ll also showcase some of the industry’s best practices and benchmark results, allowing you to see the accelerated performance GPUs have enabled on QCT POD.
[S42419] Empower AI workloads with NVIDIA AI Enterprise Based on QCT Infrastructures
Ryan Kuo, Product Manager, QCT
QCT, NVIDIA and VMware are offering a new solution to empower AI applications for enterprises in a virtualized environment. QCT’s validated design for AI includes the latest version of VMware vSphere and Tanzu combined with the NVIDIA AI Enterprise software suite on QCT servers. In addition, it features a disaggregated server architecture to offer customers higher flexibility. Join us to learn how this design can enable AI workloads with the GPU virtualization function (Multi-Instance GPU).