Yahav Biran's repositories
monotonous_id
generating monotonous id
ml-train-mix-gpu-ai-accelerator
Mix Trainium and GPU in ML training
aws-neuron-samples
Example code for AWS Neuron SDK developers building inference and training applications
aws-neuron-sdk
Powering AWS purpose-built machine learning chips. Blazing fast and cost effective, natively integrated into PyTorch and TensorFlow and integrated with your favorite AWS services
code-generator
Code generator for AWS Controllers for Kubernetes
containerized-game-servers
Demonstrate game-server related deployment methods on EKS
Craft
A simple Minecraft clone written in C using modern OpenGL (shaders).
craft-auth-disco
Implement auth protocol of yahavb/Craft
data-on-eks
DoEKS is a tool to build, deploy and scale Data Platforms on Amazon EKS
ds-resource-injector
webhook that injects daemonset pods resources.requests based on the instance size
edge_diffusion_on_eks
Deploy HuggingFace Stable Diffusion 1.5 (512x512) model Neuron and NVIDIA nodes at the edge
eks_gpu_and_trainuim_perceiver_io_training
This example shows how to produce multimodal videos with audio using the Kinetics dataset on AWS Trainium and EC2 GPU orchestrated by EKS and launched by Karpenter
gam302r
gam302r
game302-cd-test-1
game302-cd-test-1
karpenter
Karpenter is a Kubernetes Node Autoscaler built for flexibility, performance, and simplicity.
multi_chip_modal_perceiver_training
train_kinetics dataset
musical-bassoon
application-health-sample-for-aws-load-balancer-controller
optimum-neuron
Easy, fast and very cheap training and inference on AWS Trainium and Inferentia chips.
pgbouncer-fast-switchover
Adds query routing and rewriting extensions to pgbouncer
probable-octo-spoon
Playing with EKS, Fargate, and Karpenter
rds-controller
ACK service controller for Amazon Relational Database Service (RDS)
s3-sturdy-waddle
How to grant access to S3 bucket
scalable-hw-agnostic-inference
A hardware-agnostic (NVIDIA's GPUs and AWS Inferentia accelerators) deployment of text-to-image app on EKS controlled by K8s ingress in routing-time and Karpenter in scheduling-time that is scaled by KEDA.
udp-super-spoon-flooding-simu
UDP flood simulation