| title | Kata: 200 - Advanced Observability Stack | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| description | Learn complex ADR creation through multi-component evaluation, comparing TIG vs TICK stack for edge observability and monitoring... | ||||||||
| author | Edge AI Team | ||||||||
| ms.date | 2025-01-20 | ||||||||
| kata_id | adr-creation-200-advanced-observability-stack | ||||||||
| kata_difficulty | 2 | ||||||||
| kata_category |
|
||||||||
| estimated_time_minutes | 50 | ||||||||
| learning_objectives |
|
||||||||
| prerequisite_katas |
|
||||||||
| technologies |
|
||||||||
| success_criteria |
|
||||||||
| ai_coaching_level | adaptive | ||||||||
| scaffolding_level | medium-heavy | ||||||||
| hint_strategy | progressive | ||||||||
| common_pitfalls |
|
||||||||
| requires_azure_subscription | false | ||||||||
| requires_local_environment | true | ||||||||
| tags |
|
||||||||
| search_keywords |
|
You'll Learn: Create complex multi-component ADRs for sophisticated technology decisions involving multiple integrated systems and stakeholder requirements.
Prerequisites: Completion of Basic Messaging Architecture ADR kata, experience with observability concepts, understanding of monitoring stack components
Real Challenge: Your predictive maintenance system across 12 manufacturing facilities needs comprehensive observability. You must choose between TIG (Telegraf, InfluxDB, Grafana) vs TICK (Telegraf, InfluxDB, Chronograf, Kapacitor) stack to monitor 500+ edge nodes, 10,000+ sensors, and 100+ AI models with real-time analytics and automated alerting.
Your Task: Create an executive-level ADR that evaluates both observability stacks against technical, operational, and business criteria, documenting a clear recommendation with implementation roadmap.
💡 Pro Tip: Your AI coach can see your checkbox progress when you're using the local docs (npm run docs) and help you manage it! Navigate to the Learning section to access all learning resources.
Complete these prerequisite checks to ensure a successful learning experience:
Required (check these first):
- Completion of previous ADR kata (01-basic-messaging-architecture)
- Understanding of observability concepts (metrics, logs, traces)
- Access to solution ADR library templates and examples in /docs
💡 Pro Tip: Consider using specialized agents for different phases of your work - one approach for systematic research, another for professional documentation. Separating these cognitive tasks can improve quality and efficiency.
📁 Pro Tip: Task Researcher automatically saves research documents to .copilot-tracking/research/ - check there for your research files.
Before starting the tasks:
- Load the Task Researcher agent (
.github/agents/task-researcher.agent.md) for research phases - Load the ADR Creation Agent (
.github/agents/adr-creation.agent.md) — specialized mode for transforming research into professional ADR documentation with template structure, stakeholder perspective analysis, and executive-level writing guidance
Quick Validation: Verify you have completed the Basic Messaging Architecture kata and can access observability component documentation.
🤖 Want Interactive AI Coaching?
Load the Learning Kata Coach custom agent for task check-offs, progress tracking, progressive hints, and personalized guidance.
In GitHub Copilot Chat, select Learning Kata Coach custom agent and say:
I'm working on 02 - Advanced Observability Stack kata and want interactive coaching with progress tracking.
What You'll Do: Research and analyze both observability stacks' architectures, components, and integration patterns.
Steps:
-
Create research document using Task Researcher agent
- Switch to or confirm Task Researcher agent is active
- Copy and paste this exact prompt:
Help me research TIG (Telegraf, InfluxDB, Grafana) and TICK (Telegraf, InfluxDB, Chronograf, Kapacitor) observability stacks for edge computing with 500+ nodes. Focus on: - Scalability (handling 1M+ data points/minute) - Resource efficiency (edge deployment constraints) - Real-time alerting capabilities - Component integration patterns Create a comparison table with architecture analysis.- Expected result: Initial research document saved to
.copilot-tracking/research/
-
Update research document with architecture analysis
- Continue working with Task Researcher agent
- Request analysis of: TIG stack component relationships, data flow patterns, edge deployment patterns
- Request analysis of: TICK stack complete platform integration, stream processing, alerting capabilities
- Request analysis of: Performance characteristics at scale, resource footprint comparison
- Expected result: Comprehensive understanding of both architectures with scalability focus
-
Continue updating research document with edge computing analysis
- Still working with Task Researcher agent
- Request analysis of: Data collection strategies, resource utilization patterns
- Request analysis of: Storage optimization, query performance, distributed deployment considerations
- Success check: Can articulate key architectural differences and trade-offs between stacks with specific evidence
- Reflection checkpoint: Did you build a structured research document that captures technical comparisons and trade-offs?
What You'll Do: Define evaluation criteria addressing technical, operational, and business stakeholder needs.
Understanding Stakeholder Perspectives (read this first):
Different stakeholders care about different aspects of the observability stack decision:
- Technical teams (DevOps, SRE): Performance, scalability, integration complexity, troubleshooting capabilities
- Operations teams: Deployment complexity, maintenance overhead, training requirements, support availability
- Executive/Business: Total cost of ownership, implementation timeline, business continuity risk, vendor lock-in
- Security/Compliance: Data retention, access control, audit capabilities, regulatory requirements
Your ADR must address ALL these perspectives, not just technical merits.
Steps:
-
Identify technical requirements for the scenario
- Document performance requirements: 1M+ data points/minute across all sites, sub-second alerting, 2-year data retention
- Document integration needs: Multi-tenant monitoring, enterprise integration, distributed operations
- Copy and paste this prompt to Task Researcher:
What are the technical requirements for an observability stack supporting 500+ edge nodes with real-time monitoring and 1M+ data points per minute?- Expected result: Comprehensive technical requirements matrix
-
Define operational requirements and constraints
- Identify operational factors: Deployment complexity across edge locations, maintenance overhead
- Consider team factors: Configuration management, troubleshooting capabilities, team training needs
- Copy and paste this prompt:
What are the operational considerations for deploying and maintaining an observability stack across 12 distributed manufacturing facilities?- Expected result: Clear operational complexity assessment criteria
-
Establish business evaluation criteria
- Consider financial factors: Total cost of ownership, compliance requirements
- Consider strategic factors: Risk factors, implementation timeline, scalability considerations
- Copy and paste this prompt:
What business and financial criteria should guide the selection of an observability stack for enterprise manufacturing operations?- Success check: Multi-dimensional evaluation framework ready for analysis addressing technical, operational, and business perspectives
What You'll Do: Create a comprehensive ADR with detailed analysis, clear recommendation, and implementation roadmap.
Understanding "Executive-Level" ADR:
An executive-level ADR is more than just a technical comparison - it must:
- Lead with business impact: Start with why this decision matters to the organization
- Use clear, jargon-free language: Explain technical concepts in business terms
- Quantify where possible: Include specific metrics (cost, time, resource requirements)
- Address risk: Explicitly call out implementation risks and mitigation strategies
- Provide clear recommendation: No ambiguity - make a definitive choice with strong justification
- Include actionable roadmap: Executives want to know "what happens next" with specific timelines
Before starting documentation:
- Switch to ADR Creation agent (see Essential Setup above)
- Find your research document in
.copilot-tracking/research/and add it to the chat context (attach file) - Copy and paste this exact prompt:
I have completed research comparing TIG (Telegraf, InfluxDB, Grafana) and TICK (Telegraf, InfluxDB, Chronograf, Kapacitor) observability stacks.
Help me create an executive-level ADR for selecting an observability stack for edge computing infrastructure with:
- 500+ distributed edge nodes across 12 manufacturing facilities
- 1M+ data points/minute real-time monitoring
- Multi-stakeholder requirements (technical, operational, business)
Structure the ADR to address all stakeholder perspectives with clear business impact and implementation roadmap.
Steps:
-
Create structured technology comparison using your research analysis
- Work with ADR Creation agent to structure the document
- Context section using scenario from Quick Context (500+ nodes, edge constraints, 1M+ data points/minute)
- Apply evaluation criteria to both stacks with specific evidence from your research
- Include quantified analysis where possible (performance, cost, complexity)
- Document integration considerations with existing enterprise systems (e.g. identity and PKI)
- Expected result: Detailed comparative analysis supporting decision-making
-
Review your comparative analysis using Ask mode for differential perspectives
- Switch to Ask mode in GitHub Copilot Chat
- Attach your draft ADR to the chat context
- Ask for technical review: "Review this observability stack comparison for technical gaps, missing considerations, and alternative perspectives. What edge computing challenges might I have overlooked? Are there scalability concerns I should address?"
- Ask for stakeholder analysis: "What concerns might different stakeholders (operations team, security team, finance) raise about this analysis? What questions should I anticipate?"
- Refine your analysis based on the feedback and identified gaps
- Expected result: Strengthened analysis with addressed gaps and consideration of multiple perspectives
-
Document clear recommendation with multi-stakeholder justification
- Present selected stack with rationale addressing technical, operational, and business concerns
- Include risk assessment and mitigation strategies
- Address implementation complexity and timeline considerations
- Expected result: Executive-level recommendation with comprehensive justification
-
Include implementation roadmap and success metrics
- Phase-based deployment strategy minimizing operational risk
- Integration approach with existing monitoring systems
- Training plan and knowledge transfer requirements
- Success criteria: Complete ADR ready for stakeholder review and implementation approval
You've Succeeded When:
- Created comprehensive ADR comparing TIG vs TICK stacks with detailed analysis
- Used Task Researcher agent to create structured research before ADR creation
- Used Ask mode to review and strengthen analysis with differential perspectives
- Separated research phase from documentation phase for better quality outcomes
- Addressed technical, operational, and business stakeholder requirements
- Documented clear recommendation with implementation roadmap
- ADR meets executive-level quality standards for complex technology decisions
Ready for more? Try creating an ADR for a related technology decision (e.g., commercial vs open-source monitoring, hybrid cloud/edge observability) using this same process.
Core Learning Tools:
- Task Researcher Agent -
.github/agents/task-researcher.agent.mdfor structured research - ADR Creation Agent - See Essential Setup section for full path and description
- ADR Solution Library -
docs/solution-adr-library/for templates and examples
Azure Observability Documentation:
- Azure Monitor - Comprehensive monitoring and observability platform guidance
- Application Insights - Application performance monitoring integration
- Previous ADR - Reference methodology and format for consistency
- Break down complex multi-component stack evaluations into focused research phases (architecture, requirements, comparison) before synthesizing into ADR
- Use multiple AI agents strategically: Task Researcher for gathering evidence, Ask mode for challenging your assumptions, ADR Creation for professional documentation
- Address multi-stakeholder perspectives early - technical, operational, and business criteria often conflict and require explicit trade-off analysis
- Include quantified metrics wherever possible (performance numbers, cost estimates, resource requirements) to strengthen technical arguments
Issue: Complex technology comparison becomes overwhelming with too many factors
- Quick Fix: Focus on key differentiators and prioritize evaluation criteria based on specific use case requirements. Use comparison tables to structure analysis.
Issue: Uncertain which observability metrics matter most for edge scenarios
- Quick Fix: Prioritize metrics based on business impact (uptime, performance, cost). Start with must-have requirements before nice-to-have features.
Issue: Implementation roadmap unclear or too vague for stakeholder approval
- Quick Fix: Document phased deployment with specific milestones, integration points, and success criteria. Include risk mitigation strategies.
🤖 Crafted with precision by ✨Copilot following brilliant human instruction, then carefully refined by our team of discerning human reviewers.