Bright Computing will be exhibiting at the GPU Technology Conference (GTC 2017) being held May 9-11 at the San Jose Convention Center, and we expect it to be an exciting event!
This year we will be giving live demos of the new releases of Bright Cluster Manager™ and Bright OpenStack. The demos will cover the management and monitoring of GPUs in HPC, big data, and deep learning clusters, how to run deep learning workloads on clusters that combine on-premises and cloud resources in AWS and Azure, and a view of our ready-to-use Deep Learning frameworks, libraries and tools.
Join Martijn de Vries, CTO of Bright Computing, for technical session S7161 "OpenStack + AWS, HPC (aaS), and GPUs - a Pragmatic Guide" scheduled for Monday, May 8 from 9:00 - 9:50am in room 210E. This session will address some of the latest challenges and “how-to’s” in the A to Z of HPC and Big Data:
- “Why and how to do HPC (aaS)” on-premise or in the cloud.
- Creating full GPU HPC clusters from scratch and on-demand within 5-minutes – all equipped with NVIDIA DCGM and CUDA environments, and deep learning frameworks/libraries.
- Managing hybrid clouds with GPUs spanning OpenStack, AWS, and Azure.
- Automatically moving HPC user data and workloads between the private and public cloud.
- Dynamically scaling a virtualized HPC cluster, both horizontally (within private cloud) and vertically (to public cloud).
Slides are available here.
Martijn addresses some of these questions during a deep dive into the world of HPC on top of OpenStack, AWS, and Azure. He will discuss many ways OpenStack private clouds can be used for bursting HPC workloads, HPC-as-a-service, XaaS (anything-as-a-service), and creating hybrid clouds composed of on-premise private/community cloud OpenStack deployments which dynamically scales them to public clouds such as AWS.
While you are at the conference, please look for Bright Computing booth #628 to learn how our platform makes it easy for IT organizations to deliver the computing resources and support that analysts and researchers need to manage a high-performance computing cluster; or a Hadoop- or Spark-based big data environment; or to deploy OpenStack to manage resources more effectively – all within one unified management platform.
We look forward to seeing you in San Jose!