Uploaded image for project: 'OpenShift Container Platform (OCP) Strategy'
  1. OpenShift Container Platform (OCP) Strategy
  2. OCPSTRAT-2316

Confidential Clusters with remote attestation - Phase I

XMLWordPrintable

    • Product / Portfolio Work
    • OCPSTRAT-2023OpenShift Confidential Clusters
    • False
    • Hide

      None

      Show
      None
    • False
    • None
    • None
    • None
    • None
    • None
    • None
    • None

      Feature Overview (aka. Goal Summary)  

      Establish the foundational architecture and community presence for the Confidential Cluster Operator by creating the upstream repository, designing the comprehensive solution architecture, and socializing the approach through technical content. This feature delivers the architectural blueprint and community framework that enables Red Hat and ecosystem contributors to build a production-ready operator that automates confidential computing capabilities across multi-cloud OpenShift deployments.

      Goals (aka. expected user outcomes)

       

      Primary User Types/Personas:

      • OpenShift Engineering Teams: Have a clearly defined architecture and upstream repository to begin development work on the Confidential Cluster Operator
      • Community Contributors: Can discover, understand, and contribute to the confidential cluster capabilities through accessible upstream channels
      • Technical Audience (Architects, Security Engineers): Gain understanding of how OpenShift will deliver confidential computing through published technical content
      • Product Management & Field Teams: Have referenceable technical content to discuss the solution approach with customers and partners
         

      Observable Functionality:

      • A public upstream repository exists with governance, contribution guidelines, and initial project structure
      • A documented architecture defines how the operator will integrate with OpenShift, Red Hat build of Trustee, and cloud provider TEE capabilities across AWS, Azure, and GCP
      • Published blog post(s) explaining the confidential cluster approach, use cases, and technical architecture
      • Clear technical foundation enabling subsequent development features{}

      Requirements (aka. Acceptance Criteria):

      Functional Requirements:

      1. Upstream Repository Established
        • Public repository created in the appropriate upstream organization
      2. Architecture Documentation Complete
        • Architecture design document covers:
          • High-level system architecture and component interactions
          • Integration points with OpenShift installation process 
          • Attestation flow using Red Hat build of Trustee for each cloud provider
          • Node bootstrap and admission control mechanisms
          • Multi-cloud abstraction layer for TEE-specific implementations
          • Operator lifecycle (installation, upgrade, configuration)
          • Security model and trust boundaries
        • Architecture reviewed and approved by OpenShift Architecture and Security teams
        • Sequence diagrams for key workflows (initial cluster installation, node addition, attestation failure scenarios)
      3. Technical Content Published
        • Blog post published on Red Hat Developer or OpenShift blog
        • Content includes:
          • Problem statement and confidential computing value proposition
          • Solution overview (what OpenShift Confidential Clusters will deliver)
          • High-level architecture explanation

      Non-functional Requirements:

      • Maintainability: Architecture design allows for extensibility to additional cloud providers and TEE technologies
      • Security: Architecture document includes threat model and security considerations
      • Reliability: Architecture includes failure modes and recovery mechanisms for attestation failures
      • Usability: Repository and documentation use clear, consistent terminology; architecture is understandable by OpenShift engineers without deep confidential computing expertise

       

      Deployment considerations List applicable specific needs (N/A = not applicable)
      Self-managed, managed, or both Self-managed initially; architecture should consider future managed service requirements (ARO, ROSA, OSD)
      Classic (standalone cluster) Yes - primary target 
      Hosted control planes Out of scope for initial architecture; document as future consideration 
      Multi node, Compact (three node), or Single node (SNO), or all Architecture must support multi-node and compact (3-node); SNO as future consideration (document constraints)
      Connected / Restricted Network Both - architecture must account for attestation service connectivity in restricted networks (document requirements)
      Architectures, e.g. x86_x64, ARM (aarch64), IBM Power (ppc64le), and IBM Z (s390x) x86_x64 initially (all three cloud providers support confidential VMs on x86); ARM as future consideration
      Operator compatibility Define dependencies on platform operators (machine-config, cluster-version); document OLM integration approach
      Backport needed (list applicable versions) N/A - new capability for future OpenShift release 
      UI need (e.g. OpenShift Console, dynamic plugin, OCM) Not required for this feature; document future UI requirements in architecture (console integration for attestation status visibility)
      Other (please specify) Cloud provider prerequisites (TEE-enabled instance types, specific CPU generations) must be documented

      Out of Scope

      • Actual operator implementation code (covered by subsequent development features)
      • Workload-level confidentiality (e.g., confidential containers for individual pods - different solution)
      • Key management and encryption key provisioning (architectural dependency, not delivered in this feature)
      • Hosted control plane support (HyperShift integration deferred to future)
      • Performance benchmarking and optimization (covered in later features)
      • Customer-facing documentation (will be created as part of GA documentation effort)
      • Multi-cluster or fleet management scenarios

       

              mak.redhat.com Marcos Entenza Garcia
              mak.redhat.com Marcos Entenza Garcia
              None
              Clement Verna, Nitesh Narayan Lal
              Timothée Ravier Timothée Ravier
              Yalan Zhang Yalan Zhang
              Avani Bhatt Avani Bhatt
              Kyle Walker Kyle Walker
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

                Created:
                Updated: