1 of 24

Introduction to

TAG Infrastructure

Dylan Page, Lambda.ai

Kashif Khan, Ericsson Software Technology

#KubeCon #CloudNativeCon

2 of 24

TAG Reboot

Technical Advisory Groups (TAGs)

TAG Developer Experience

TAG Workloads Foundation

TAG Infrastructure

TAG Operational Resilience

TAG Security and Compliance

Subprojects

Initiatives

Project Reviews Subproject

Contributor Strategy Subproject

Artificial Intelligence Initiatives

Technical Oversight Committee (TOC)

  • Consolidated TAGs from 8 to 5
  • Replace Working Groups with Subprojects and Initiatives
  • Alignment with K8s working group concept
  • Any previous Working Group may apply to be a new Subproject or Initiative

3 of 24

Agenda

  • How TAGs are restructuring
    • Motivation for the change
    • Previous vs Current structure

  • What does this TAG do and how we help projects
    • Who we are
    • Charter Scope
      • Example CNCF projects
    • How we can help?
      • Subproject Reviews (Sandbox -> Graduation)
    • Initiatives
      • Infrastructure Lifecycle
      • CNCF Storage Landscape Whitepaper V3
      • Data Storage and Cloud Native AI Whitepaper

  • Community
    • How to join and be part of the community

4 of 24

TAG Reboot References

TAG Reboot TOC GitHub Issue

TAG Reboot Presentation Slides

5 of 24

Who we are

Co-Chairs

Dylan Page

Kashif Khan

Xing Yang

Tech Leads

Alexa Griffith

Antonio Ojea

Bruno Schaatsbergen

Nicholas Jackson

Zhonghu Xu

TOC Liaison: Ricardo Rocha & Karena Angell

We are a diverse community of developers and end-users of Cloud Native technologies with a focus on Data, Storage, Network, DNS, Compute, Service Mesh, Infrastructure-as-Code, Edge, Sovereignty, and Load Balancing.

6 of 24

Charter: Mission & Focus

  • Defines & advances cloud-native infrastructure practices
  • Ensures scalable, resilient, secure, performant systems
  • Addresses infra challenges faced by adopters
  • Aligned with CNCF technical vision & TOC charter

7 of 24

Charter: Core Technical Domains

  • Data: logs, metrics, configs, user data
  • Storage: block, file, object, DB, caching, messaging
  • Networking: DNS, gateways, service mesh, policy, observability
  • Compute: runtimes, isolation, accelerators (GPU/DPU/TPU)
  • Infra Management: IaC, orchestration, drift, compliance
  • Edge & Sovereignty: distributed, regulated, multi-cluster

8 of 24

Charter: Core Technical Domains

  • Data: logs, metrics, configs, user data
  • Storage: block, file, object, DB, caching, messaging
  • Networking: DNS, gateways, service mesh, policy, observability
  • Compute: runtimes, isolation, accelerators (GPU/DPU/TPU)
  • Infra Management: IaC, orchestration, drift, compliance
  • Edge & Sovereignty: distributed, regulated, multi-cluster

9 of 24

Charter: Core Technical Domains

  • Data: logs, metrics, configs, user data
  • Storage: block, file, object, DB, caching, messaging
  • Networking: DNS, gateways, service mesh, policy, observability
  • Compute: runtimes, isolation, accelerators (GPU/DPU/TPU)
  • Infra Management: IaC, orchestration, drift, compliance
  • Edge & Sovereignty: distributed, regulated, multi-cluster

10 of 24

Charter: Core Technical Domains

  • Data: logs, metrics, configs, user data
  • Storage: block, file, object, DB, caching, messaging
  • Networking: DNS, gateways, service mesh, policy, observability
  • Compute: runtimes, isolation, accelerators (GPU/DPU/TPU)
  • Infra Management: IaC, orchestration, drift, compliance
  • Edge & Sovereignty: distributed, regulated, multi-cluster

11 of 24

Charter: Core Technical Domains

  • Data: logs, metrics, configs, user data
  • Storage: block, file, object, DB, caching, messaging
  • Networking: DNS, gateways, service mesh, policy, observability
  • Compute: runtimes, isolation, accelerators (GPU/DPU/TPU)
  • Infra Management: IaC, orchestration, drift, compliance
  • Edge & Sovereignty: distributed, regulated, multi-cluster

12 of 24

Charter: Core Technical Domains

  • Data: logs, metrics, configs, user data
  • Storage: block, file, object, DB, caching, messaging
  • Networking: DNS, gateways, service mesh, policy, observability
  • Compute: runtimes, isolation, accelerators (GPU/DPU/TPU)
  • Infra Management: IaC, orchestration, drift, compliance
  • Edge & Sovereignty: distributed, regulated, multi-cluster

13 of 24

Charter: Core Technical Domains

  • Data: logs, metrics, configs, user data
  • Storage: block, file, object, DB, caching, messaging
  • Networking: DNS, gateways, service mesh, policy, observability
  • Compute: runtimes, isolation, accelerators (GPU/DPU/TPU)
  • Infra Management: IaC, orchestration, drift, compliance
  • Edge & Sovereignty: distributed, regulated, multi-cluster

14 of 24

Charter: Deliverables & Success Criteria

  • Outputs: frameworks, whitepapers, best practices
  • Initiatives: focused, time-bound research or assessments
  • Subprojects: ongoing stewardship and services
  • Success: active workstreams, community growth, adoption

15 of 24

Charter: Coordination & Alignment

  • Work with CNCF Projects, TAGs, and TOC subprojects
  • Drive ecosystem-wide alignment on infra standards
  • Enable cross-TAG collaboration for cohesive stacks
  • Advance TOC’s problem-centric technical vision

16 of 24

Initiative: CNCF Storage Landscape V3

  • Initiative issue: CNCF Storage Landscape Whitepaper V3
  • Definition of the attributes of a storage system
  • Definition of the layers in a storage solution with a focus on terminology and how they impact the attributes
  • Definition of the data access interfaces in terms of volumes and application APIs
  • Definition of the management interfaces
  • WIP Whitepaper V3

17 of 24

Initiative: Data Storage in Cloud Native AI

  • Initiative issue: Data Storage in Cloud Native AI
  • Describe characteristics of AI/ML workloads and their implications for data storage.
  • Describe patterns and trends in data storage for cloud native AI
    • Data warehouses, data lakes, and data lake houses
    • Data cache and data locality
    • Vector databases
    • Block, file, and object storage
    • Data mesh and data fabric
  • Describe the storage requirements and usage patterns in the AI lifecycle including training, inference, etc.
  • WIP Whitepaper

18 of 24

Initiative: Infrastructure Lifecycle

  • Initiative Issue: Infrastructure Lifecycle #1631
  • End-to-end infra lifecycle: design → deploy → operate → update → decommission
  • Guided by cloud-native principles: secure, resilient, observable, manageable
  • Applies to public, private, and hybrid cloud environments
  • Produces framework for operational best practices
  • Enables automation, drift detection, policy enforcement, and sustainability
  • WIP Whitepaper ?

19 of 24

Ways to Contribute

    • Contributors to supporting existing CNCF projects!!!
    • Contributors to Initiatives
    • Contributors to TOC Subproject Project Reviews
      • Domain Technical Reviews
    • Refining of the TAG scope
      • Are there gaps?
    • Identifying industry trends
      • Upcoming trends
      • Best practices
    • Community advocates (social media, talks, etc)

20 of 24

Contributor Ladder

21 of 24

Community Links

CNCF Slack #tag-infrastructure

TAG Infrastructure LFX Meetings

22 of 24

Thank You

23 of 24

CNCF Storage Projects

Graduated

Incubating

24 of 24