Ask an OpenShift Admin (Ep 107) Red Hat OpenShift Data Science

Joined by Robert Lundberg from the Red Hat OpenShift Data Science (RHODS) team to discuss the RHODS platform! Check out the stream here: https://www.youtube.com/live/DDvkMj-czFU?feature=share

Transcript

1. Presentation title should not
   exceed two lines
   1
   MLOps platform for artificial intelligence
   / machine learning (AI/ML) use cases
   Overview of Red Hat
   OpenShift Data Science
   Robert Lundberg
   Senior Architect at OpenShift AI
   

   View Slide

2. 2
   Source: https://proceedings.neurips.cc/paper/2015/file/86df7dcfd896fcaf2674f757a2463eba-Paper.pdf
   Hidden Technical Debt in Machine Learning Systems
   Model code is one component of a larger system
   "Only a small fraction of real-world
   ML systems is composed of the
   ML code…The required
   surrounding infrastructure is
   vast and complex."
   "Developing and deploying
   ML systems is relatively fast
   and cheap, but maintaining
   them over time is difficult
   and expensive"
   Developing and deploying ML is fast and cheap >>
   configuration
   data collection
   data
   verification
   machine
   resource
   management
   serving
   infrastructure
   monitoring
   analysis tools
   process
   management
   feature extraction
   (Adapted from Sculley et al., "Hidden Technical Debt in Machine Learning Systems." NIPS 2015
   model code
   

   View Slide

3. Why containers, Kubernetes and Red Hat OpenShift for Data Science >>
   3
   https://kubernetes.io/
   https://www.redhat.com/en/technologies/cloud-computing/openshift/red-hat-openshift-kubernetes
   https://aws.amazon.com/blogs/opensource/why-use-docker-containers-for-machine-learning-development/
   Why Kubernetes?
   ● Automated rollouts and rollbacks
   ● Self-healing
   ● Service discovery and load balancing
   ● Horizontal scaling
   ● Designed for extensibility
   Why OpenShift?
   ● Self Service Model
   ● Web UI based Workflows
   ● Metrics and Monitoring
   ● Real-Time, Batch and Streaming Support
   ● Users can Focus on Data Science
   ● Zero Trust Security Model
   ● GPU Support
   ● Cloud and Platform Agnostic
   Why containers?
   ● Fewer resources
   ● Environment isolation
   ● Quick deployment
   ● Quick startup/shutdown
   ● Encapsulation and portability
   ● Reusability
   ● Reproducible
   Why containers, K8s, RHOCP for Data Science?
   Kubernetes
   ...
   Red Hat
   OpenShift
   Production
   Ready
   IMAG
   E
   CONTAINE
   R
   IMAGE
   Image registry
   IMAG
   E
   IMAG
   E
   IMAG
   E
   IMAG
   E
   IMAG
   E
   [ containers, Kubernetes, Red Hat Openshift ]
   

   View Slide

4. Upstream code enhanced with operational excellence
   Open Data Hub: https://opendatahub.io
   Overview of Red Hat OpenShift Data Science
   4
   Red Hat OpenShift Data Science
   Open Data Hub
   Community driven upstream meta-project demonstrating AI/ML platform
   on Red Hat OpenShift comprised of open source projects
   Red Hat OpenShift Data Science: cloud service
   Subset of Open Data Hub delivered as a cloud service on Red Hat
   OpenShift Managed services with optional ISV offerings
   Red Hat OpenShift Data Science: self managed
   Fast moving software stream mirrors the release frequency and capabilities of the
   cloud service delivered in a self-managed offering for on-premise OpenShift
   customers
   

   View Slide

5. 5
   Red Hat OpenShift Data Science
   Hybrid MLOps platform
   Model development
   Conduct exploratory data science in JupyterLab with access
   to core AI / ML libraries and frameworks including TensorFlow
   and PyTorch using our notebook images or your own.
   Collaborate within a common
   platform to bring IT, data science,
   and app dev teams together
   Model serving & monitoring
   Deploy models across any cloud, fully managed, and
   self-managed OpenShift footprint and centrally monitor their
   performance.
   Lifecycle Management
   Create repeatable data science pipelines for model training
   and validation and integrate them with devops pipelines for
   delivery of models across your enterprise.
   Increased capabilities / collaboration
   Create projects and share them across teams. Combine Red
   Hat components, open source software, and ISV certified
   software.
   Now available as fully managed cloud
   service or traditional software product
   on-prem or in the cloud!
   

   View Slide

6. Gather and prepare data Integrate models in app dev
   Model monitoring
   and management
   Develop model
   6
   Overview of Red Hat OpenShift Data Science
   Open hybrid cloud platform with self service capabilities
   Accelerators
   Cloud infrastructure
   Red Hat OpenShift Data Science
   Red Hat OpenShift Service on
   Amazon Web Services
   Retrain models
   ISV managed cloud
   services
   Red Hat software
   and cloud services
   Red Hat on premise
   and cloud platform
   Customer managed
   ISV software
   Cloud service and self-managed components
   model serving
   Streams data science pipelines model monitoring
   

   View Slide

7. Demo
   

   View Slide

8. linkedin.com/company/red-hat
   youtube.com/user/RedHatVideos
   facebook.com/redhatinc
   twitter.com/RedHat
   Red Hat is the world’s leading provider of
   enterprise open source software solutions.
   Award-winning support, training, and consulting
   services make
   Red Hat a trusted adviser to the Fortune 500.
   Thank you
   8
   

   View Slide