Amazon Elastic Kubernetes Service (EKS) vs. Amazon SageMaker

Well suited for microservices architecture but can be a bit costly if less number of microservices or monolithic architecture hosted to be hosted on containers. Use of hybrid cluster instances also works well using both normal and fargate instances. Also the integration of audit and diagnostic logs of master nodes helps to reduce the unwanted access related issues.

Incentivized

Amazon Sagemaker suits well in areas of data science and Machine learnings where medium to high-volume data is to be used for analysis. For a lean and platform agnostic deployment, it provides kubernetes integration to containerize the solution and deploy on any platform. It is one of the best solution for technical users for training Machine Learning models.

Incentivized

• Upgrade the kubernetes clusters to the latest version with a single click
• Auto scaling policies to automatically scale the nodes
• Detailed logs and events on the cluster within the EKS clusters portal, cloudwatch logs and metrics

Incentivized

• SageMaker is useful as a managed Jupyter notebook server. Using the notebook instances' IAM roles to grant access to private S3 buckets and other AWS resources is great. Using SageMaker's lifecycle scripts and AWS Secrets Manager to inject connection strings and other secrets is great.
• SageMaker is good at serving models. The interface it provides is often clunky, but a managed, auto-scaling model server is powerful.
• SageMaker is opinionated about versioning machine learning models and useful if you agree with its opinions.

Incentivized

• AWSIAM integration with Kubernetes RBAC could be better.
• Enabling some add-ons like service mesh, and monitoring will be nice instead of having to install them yourself after the creation of the cluster.
• EKS bootstrap time could be faster ...

Incentivized

• Searching and descriptions can be easier to read and interpret.
• Training modules and customer service training representative could make on boarding employees easier.

Incentivized

Cluster maintanence is reduced, easier to deploy resources, great observability insights

Incentivized

It feels like AWS is behind the EKS race, the only advantage I'm able to see right now is the support of IPv6, however, trying to promote AWS alternatives that are different from the market and more like a vendor locking solutions like ECS/Fargate have kept AWS behind and focusing on the wrong things. EKS needs to really improve its integration with the Kubernetes ecosystem and have an enterprise solution for monitoring, backups, and service mesh.

Incentivized

We have not invested in another machine learning software at this time and so far this has proved very successful with our machine learning teams. As mentioned, I am training these individuals simply on the fundamentals of the software and using it/customizing it for their needs. It has been very easy to do this and has gotten great reviews across the organization so far.

Incentivized

• Migrating all our workloads from ec2 VMs to containers running in Kubernetes has been a huge improvement for the management and resilience of our Infrastructure.
• EKS Upgrade process to a new version seems to be taking very long ....
• EKS creation time usually takes over 10 minutes in us-east-1, we would like faster creation times to be under 5 minutes.

Incentivized

• Using SageMaker, we can truly implement 'fail early, learn fast,' using an on-demand server for training.
• It also saves your money from investing in a physical server for very rare use.
• However, the pricing is high, but it will cost you only for what you use.

Incentivized