Amazon EKS introduces Beta version of Amazon FSx for Lustre CSI Driver

The Amazon FSx for Lustre CSI driver has achieved beta status and is now sustained by Amazon Elastic Kubernetes Service (EKS). The CSI driver composes it elementary to design and utilise FSx for Lustre high functioning file systems with packages on EKS and autonomous Kubernetes clumps functioning on AWS. Amazon FSx for Lustre is an entirely counseled, high functioned file system enhanced for workloads. With FSx for Lustre, one can expeditiously and conveniently swirl up a high-performance file system associated with the S3 data archives, and access S3 objects as files. With the FSx for Lustre CSI driver, one can emphatically equip and mount an FSx for Lustre file system to packages, so that the containerised workloads can constitutionally access and operate data reserved in the file system or S3 data repository. One can utilise the CSI driver to mount and distribute the FSx file system over multiple cases from different nodes. Containerised operations that need high functioning storage can assist from using the FSx for the Lustre file system with the CSI driver. Such features comprise distributed machine learning workloads on the structures such as Tensor-flow and PyTorch, and media refining workloads. Administering high-performance file systems requires specially designed expertise and bureaucratic overhead, required to arrange the storage servers and tune convoluted performance criterion.

Amazon EKS gives assistance to G4 Instance

Amazon Elastic Kubernetes Service (Amazon EKS) is an open source and entirely organised service which makes it simple for you to execute Kubernetes on AWS excluding requirements to stand up or retain your Kubernetes control plane. Kubernetes system used for automating the deployment, scaling, and management of containerised applications. Now this Amazon Elastic Kubernetes Service (Amazon EKS) give assistance to adding Amazon EC2 G4 instances as worker nodes to each cluster in AWS Regions where G4 instance is obtainable. Instance G4 is the latest EC2 GPU-equipped (Graphics Processing Unit) instance and gives the commercially most cost-effective GPU platform for deploying machine learning models in production and graphics-intensive applications. G4 instances are optimized for machine learning application deployments (inference), like object detection, image classification, recommendation engines, automated speech recognition, and language translation which require access to low level GPU software libraries. To get started with G4 instance in your cluster, first you require to upgrade to 1.5.4 version of AWS VPC CNI plugin. Refer documentation to know the upgrading directions of CNI plugin. To add GPU instance to your cluster, read Amazon EKS documentation. And to know further about G4 instances, go through G4 page.

Now Amazon EFS CSI Driver Supported by Amazon EKS

Amazon Elastic File System (Amazon EFS) offers an easy, scalable and completely organised elastic network file system NFS for utilization with AWS Cloud Services and on-premises resources. And Amazon EFS CSI (Container Storage Interface) Driver offers a CSI interface which permits Amazon EKS clusters to control the lifecycle of Amazon EFS file systems. And now this CSI driver has introduced beta version and supported by Amazon EKS. This EFS CSI driver is easy to configure elastic file storage for EKS and self-managed kubernetes clusters working on AWS with the help of basic Kubernetes interfaces. Applications executing in Kubernetes can utilize EFS file systems to share data within pods in a scale-out group, or with other applications working inside or outside of Kubernetes. Besides, Amazon EFS support Kubernetes applications be extensively obtainable cause each data written to EFS is corresponded to several AWS Availability Zones. Refer documentation to design a Kubernetes StorageClass and a PersistentVolumeClaim to provision, then reference the PersistentVolumeClaim from the pods that use it. To get more details, go through Amazon EKS and Amazon EFS.

Cluster Tagging is now supported by Amazon EKS

Amazon Elastic Kubernetes Service (Amazon EKS) is an open-source organised service which helps to automate deployment, scaling, and management of containerized applications and simple to execute Kubernetes on AWS excluding to stand up or keep your individual Kubernetes control plane. Amazon EKS executes Kubernetes control plane instances across different Availability Zones to certify high availability. This automatically identifies and substitutes harmful control plane instances, and provides automated version upgrades and patching for them. Earlier, it was hard to manage access to handle the EKS API for discrete clusters cause tagging was not available for Amazon EKS clusters and you could not add metadata that specified the motive of the cluster for reporting reasons. Now you can add AWS tags to your Amazon EKS clusters. This makes it simple to manage access to the EKS API for handling your clusters. Also you can add custom tags to each Amazon EKS cluster. These tags can be used to control access to the EKS API for handling the cluster with the help of IAM. Now, cluster tagging is supported for every new and existing EKS clusters through the AWS API, Console, and SDKs. Support for tagging clusters using CloudFormation and the eksctl command line tool is coming soon.

Fluent Bit Plugins for AWS Introduced By Amazon ECS

Amazon Elastic Container Service (Amazon ECS) is a highly scalable, high-performance container orchestration service which helps Docker containers and permits you to directly execute and scale containerized applications on AWS. Amazon ECS removes the necessity for you to install and run your personal container orchestration software, manage and scale a cluster of virtual machines, or schedule containers on those virtual machines. AWS Container Services introduces AWS Fluent Bit, a container image pre-installed with Amazon CloudWatch and Amazon Kinesis Data Firehose plugins which guides users route container logs to several targets like CloudWatch, Amazon S3, Amazon Redshift, and Amazon Elasticsearch Service. Fluent Bit is an open source and multi-platform Log Processor and Forwarder that enables you to gather data and logs from multiple sources, merge and transfer them to different places. It's completely compatible with Docker and Kubernetes environments. With the help of recently introduced AWS Fluent Bit image, you can send container logs to the range of services for logs storage and analytics at AWS. The blog about how to use the Fluent Bit image on both Amazon ECS and Amazon EKS will help you to learn further.