GPUs do more than move shapes on a gamer’s screen - they increasingly move self-driving cars and 5G packets, running on Kubernetes. Pramod Ramarao is a Product Manager at NVIDIA, and joins your hosts to talk about accelerators, containers, drivers, machine learning and more.
Do you have something cool to share? Some questions? Let us know:
- web: kubernetespodcast.com
- mail: kubernetespodcast@google.com
- twitter: @kubernetespod
Chatter of the week
- Printer networking
- Adam wants software-defined faucets
- Glowing LED faucet - where does the electricity come from?
- Faucet, a SDN controller
News of the week
- Google Cloud launches Application Manager for GKE in Beta
- GKE Surge Upgrades GA
- GKE Node Locations GA
- Anthos Ready Storage qualification
- Kafka disaster recovery with Supertubes from Banzai Cloud
- StackRox’s State of Container and Kubernetes Security report
- Cilium 1.7
- Convox launches multi-cloud
- Pangolin, an experimental Kubernetes autoscaler by Damian Peckett
- Dell/EMC rack-in-a-box
- Platform9 now distributed by Promark
- GKE security updates & defense-in-depth strategies
- Best practices for enterprise multi-tenancy with GKE
- Andrew Allbright contributes to Minikube
- Kubernetes Contributor Summit schedule announced
- That discount code again again again: KCEUGKP15
Links from the interview
- NVIDIA
- Graphics Processing Unit (GPU)
- Differences between CPU and GPU
- The math co-processor
- General-purpose computing on GPUs (commonly known as GPGPU)
- CUDA, with a C
- OpenGL and Vulkan, with a K
- Kubernetes on NVIDIA GPUs
- Device plugins for Kubernetes and scheduling GPUs
- NDC Hub for drivers and containers
- NVIDIA EGX for Edge computing with Kubernetes
- Deep Learning Training vs Inferencing
- NVIDIA GPU operator
- Pramod Ramarao
Original Source: https://kubernetespodcast.com/episode/092-nvidia/