Sign up for our newsletter and get the latest HPC news and analysis.
Send me information from insideHPC:


insideHPC Guide to QCT Platform-on-Demand Designed for Converged Workloads – Part 4

Not too long ago, building a converged HPC/AI environment – with two domains: High Performance Computing (HPC) and Artificial Intelligence (AI) – would require spending a lot of money on proprietary systems and software with the hope that it would scale as business demands changed.

In this insideHPC technology guide, “insideHPC Guide to QCT Platform-on-Demand Designed for Converged Workloads,”as we’ll see, by relying on open source software and the latest high performance/low cost system architectures, it is possible to build scalable hybrid on-premises solutions that satisfy the needs of converged HPC/AI workloads while being robust and easily manageable.

Deep Learning with QCT POD

Containerization and prepared frameworks also enable rapid deployment  of AI/ML/DL workloads on QCT POD.

QCT POD management tools efficiently allocate Docker and Kubernetes containerized resources, giving users  access to an NVIDIA GPU enabled JupyterHub container for their Deep Learning development.

QCT POD fully integrates JupyterLab, providing users with a default web-based environment workspace where they can directly upload their datasets. Included are:

  • Jupyter notebook, code editor, and command-line tools, made available through a single webpage.
  • NVIDIA GPU-compatible Deep Learning frameworks such as Tensorflow, Pytorch, and others, as well as  Anaconda for data science Python developers. All these frameworks are pre-installed and immediately  available for users to manage their diverse projects.

In this example, we’ve prepared ResNet50 with Pytorch and performed the training using two NVIDIA®  Tesla® V100 GPUs. In a future roadmap, QCT will also provide NVIDIA® Tesla® A100 GPUs to accelerate  performance on deep learning training and inference. Users can pull pretrained models from NVIDIA® NGC™ (NGC), and the whole process—allocating resources, preparing the data model, data preprocessing,  training—can be performed in a single workspace on the QCT POD.

QCT POD provides a comprehensive Deep Learning development environment that is ready for use over a widely diverse set of workloads.

Summary

QCT and their partners are dedicated to providing best in class solutions using modular designs and integration of NVIDIA HPC/DL SDK to meet the most demanding HPC, AI, and data science workloads.

The adoption of NVIDIA GPUs allows users to accelerate performance for many varied workloads. With QCT  POD management, storage, and compute building blocks, NVIDIA GPUs provide a powerful environment for  diverse HPC, Machine Learning, Cloud Services, and Edge Computing workloads. Designed to meet the  demands from each domain and industry, QCT POD delivers the unique hardware and software combinations configured to solve these challenges and achieve best performance on customer applications and workloads.

The QCT POD, with its pre-installed frameworks and pre-compiled libraries, support for containers and  advanced job/resource management tools, and embracing the latest open source technologies, offers the  best in class solution for a company’s converging HPC and AI workflows.

About QCT

Quanta Cloud Technology (QCT) is a global data center solution provider. We combine the efficiency of  hyperscale hardware with infrastructure software from a diversity of industry leaders to solve next-generation  data center design and operation challenges. QCT serves cloud service providers, telecoms and  enterprises running public, hybrid and private clouds.

Product lines include hyper-converged and software-defined data center solutions as well as servers, storages, switches, integrated racks with a diverse ecosystem of hardware components and software partners. QCT designs, manufactures, integrates and services cutting-edge offerings via its global network. The parent  of QCT is Quanta Computer, Inc., a Fortune Global 500 corporation.

To learn more about the QCT, please visit go.qct.io.

Over the past few weeks we’ve explored QCT’s Platform-on-Demand designed for converged workloads:

Download the complete insideHPC Guide to QCT Platform-on-Demand Designed for Converged Workloads courtesy of QCT.

Leave a Comment

*

Resource Links: