Vara Bonthu

Batch inference is more than running a model over a large dataset. In this interactive chalk talk, we build an end-to-end batch inference architecture on Amazon EKS one component at a time.

Kubernetes has become a standard for organizations looking to modernize their application portfolios. AWS developed Amazon EKS to make Kubernetes more accessible to organizations of all sizes, helping them focus on what matters most for their businesses.

Running Apache Spark on Amazon EKS delivers scalability, but poor defaults can drive up costs and slow pipelines. This ChalkTalk highlights the top Spark performance challenges and production-proven fixes.

Organizations use their data to make better decisions and build innovative experiences for their customers. They are also looking for modern methods of processing large volumes of data to gain greater insight.

Organizations are adopting Kubernetes to run their workloads for flexibility, scale, and cost efficiency. Deploying production-grade workloads can be simplified through the use of operator patterns.

Are you eager to build and manage large-scale Spark clusters on Kubernetes for powerful data processing? Whether you are starting from scratch or considering migrating Spark workloads from existing Hadoop clusters to Kubernetes, the challenges of configuring storage, compute, networking, and...

In this builders’ session, learn about using Amazon EKS Blueprints for Terraform to configure and deploy “batteries-included” Amazon EKS clusters. Dive deep into EKS Blueprints code and learn to deploy security-compliant EKS clusters with managed node groups and AWS Fargate profiles.