Fitness Functions Workflows¶

Step-by-step procedures for defining, implementing, and using fitness functions.

Workflow Overview¶

flowchart LR
    subgraph "Fitness Function Lifecycle"
        W1[Identify Quality<br/>Attributes] --> W2[Define Fitness<br/>Functions]
        W2 --> W3[Establish<br/>Baselines]
        W3 --> W4[Set<br/>Targets]
        W4 --> W5[Implement<br/>Measurement]
        W5 --> W6[Monitor &<br/>Report]
        W6 --> W7[Review &<br/>Adapt]
        W7 --> W2
    end

Step	Activity	Key Output
1	Identify Quality Attributes	Prioritized attribute list
2	Define Fitness Functions	Fitness function specifications
3	Establish Baselines	Current state measurements
4	Set Targets	Target values with rationale
5	Implement Measurement	Automated measurement pipeline
6	Monitor & Report	Dashboards, alerts, reports
7	Review & Adapt	Updated functions and targets

Step 1: Identify Quality Attributes¶

Purpose¶

Determine which architectural characteristics matter most for the system.

Activities¶

1.1 Stakeholder Analysis¶

Actions: 1. Identify key stakeholders (business, technical, operations) 2. Gather quality requirements from each 3. Document explicitly stated requirements 4. Infer implicit requirements from context 5. Consolidate and deduplicate

Questions to Ask: | Stakeholder | Key Questions | |-------------|---------------| | Business | What happens if system is slow? Down? Breached? | | Development | What makes code hard to change? What causes bugs? | | Operations | What causes incidents? What's hard to monitor? | | Security | What's the threat model? Compliance requirements? |

1.2 Context Analysis¶

Actions: 1. Review system criticality (revenue impact, user impact) 2. Identify growth expectations 3. Assess regulatory environment 4. Understand competitive pressures 5. Document system lifecycle stage

Context Factors: | Factor | Impact on Attributes | |--------|---------------------| | High traffic growth | Prioritize scalability | | Sensitive data | Prioritize security | | Frequent changes | Prioritize maintainability | | Real-time requirements | Prioritize performance | | Regulated industry | Prioritize compliance, auditability |

1.3 Attribute Prioritization¶

Actions: 1. List candidate quality attributes 2. Score each by business impact (1-5) 3. Score each by technical risk (1-5) 4. Calculate priority = impact × risk 5. Select top 5-7 attributes

Prioritization Matrix:

                    Business Impact
                    Low         High
Technical   High    [ Monitor ] [ Critical ]
Risk        Low     [ Ignore  ] [ Important ]

Output¶

Prioritized list of 5-7 quality attributes
Rationale for each priority
Stakeholder sign-off

Step 2: Define Fitness Functions¶

Purpose¶

Create specific, measurable fitness functions for each priority attribute.

Activities¶

2.1 Function Specification¶

For each quality attribute:

Name the fitness function clearly
Define what it measures
Specify the measurement method
Determine trigger (when to measure)
Set initial target (will refine after baseline)

Specification Format:

name: Response Time P95
attribute: Performance
description: 95th percentile response time for API requests
measurement:
  method: APM instrumentation
  tool: Datadog / New Relic / Custom
  data_source: Request traces
trigger:
  type: continuous
  frequency: Every request
target:
  value: < 200ms
  rationale: User experience research threshold
owner: Platform Team

2.2 Function Classification¶

Actions: 1. Classify by scope (atomic/holistic/composite) 2. Classify by trigger (continuous/triggered/temporal) 3. Classify by automation level 4. Identify dependencies between functions 5. Group related functions

Classification Table: | Function | Scope | Trigger | Automation | Dependencies | |----------|-------|---------|------------|--------------| | Response Time P95 | Atomic | Continuous | Automated | None | | Security Score | Composite | Temporal | Semi-automated | Vuln scan, patch status | | Architecture Health | Holistic | Temporal | Manual | Multiple metrics |

2.3 Composite Function Design¶

For composite functions:

Identify component functions
Assign weights based on importance
Define aggregation method
Handle missing components gracefully
Document calculation formula

Example:

Security Score = (
    0.4 × Vulnerability Score +
    0.3 × Patch Currency Score +
    0.2 × Auth Health Score +
    0.1 × Dependency Age Score
)

Output¶

Fitness function specifications (one per function)
Classification matrix
Composite function formulas

Step 3: Establish Baselines¶

Purpose¶

Measure current state before setting targets.

Activities¶

3.1 Baseline Collection¶

Actions: 1. Implement measurement for each function 2. Collect data over representative period 3. Calculate statistical baselines (mean, p50, p95) 4. Document measurement conditions 5. Identify anomalies and exclude if justified

Collection Period Guidelines: | Function Type | Minimum Period | Notes | |---------------|----------------|-------| | Performance | 1 week | Include peak and off-peak | | Reliability | 1 month | Capture incident patterns | | Deployment | 1 month | Capture release cycles | | Security | Point-in-time | Snapshot is valid |

3.2 Baseline Analysis¶

Actions: 1. Review baseline values 2. Compare to industry benchmarks 3. Identify obvious improvement areas 4. Note constraints and limitations 5. Document baseline context

Analysis Questions: - Is baseline representative of normal operation? - Are there seasonal or cyclical patterns? - What external factors might affect measurements? - How does baseline compare to peers/benchmarks?

3.3 Baseline Documentation¶

Document for each function:

function: Response Time P95
baseline:
  value: 450ms
  measured: 2024-01-15 to 2024-01-22
  conditions: Normal load, no incidents
  sample_size: 2.3M requests
  confidence: High
comparison:
  industry_benchmark: 200ms
  gap: 125% above benchmark
notes: |
  Current implementation has N+1 query issues
  Database connection pooling not optimized

Output¶

Baseline values for all fitness functions
Measurement conditions documentation
Gap analysis vs benchmarks

Step 4: Set Targets¶

Purpose¶

Define achievable, meaningful target values.

Activities¶

4.1 Target Setting¶

Actions: 1. Review baseline and gap analysis 2. Consider business requirements 3. Assess improvement feasibility 4. Set short-term targets (3-6 months) 5. Set long-term targets (1-2 years)

Target Setting Principles: | Principle | Description | |-----------|-------------| | Achievable | Can be reached with reasonable effort | | Meaningful | Materially improves user/business outcome | | Measurable | Can be objectively verified | | Time-bound | Has a target date | | Incremental | Steps toward long-term goal |

4.2 Target Justification¶

For each target, document:

Current baseline value
Proposed target value
Business benefit of achieving target
Technical approach to achieve
Risks to achieving target
Cost/effort estimate

Example:

function: Response Time P95
current_baseline: 450ms
target: 200ms
target_date: 2024-06-30
justification: |
  Research shows user engagement drops 7% for
  each 100ms latency above 200ms.
approach:
  - Fix N+1 queries (Q1)
  - Optimize connection pooling (Q1)
  - Add caching layer (Q2)
risks:
  - Cache invalidation complexity
  - Database schema changes needed
effort: Medium (2 engineers, 1 quarter)

4.3 Target Thresholds¶

Define multiple thresholds:

Threshold	Purpose	Example
Target	Desired state	< 200ms
Warning	Degradation alert	> 300ms
Critical	Immediate action	> 500ms

Output¶

Target values with justification
Achievement timeline
Threshold definitions (target/warning/critical)

Step 5: Implement Measurement¶

Purpose¶

Create automated, reliable measurement infrastructure.

Activities¶

5.1 Measurement Design¶

Actions: 1. Select appropriate tools per function 2. Design data collection pipeline 3. Define data storage/retention 4. Plan for measurement accuracy 5. Document measurement architecture

Tool Selection by Category: | Category | Tools | |----------|-------| | Performance | APM (Datadog, New Relic, Dynatrace) | | Reliability | Synthetic monitoring, error tracking | | Security | SAST/DAST scanners, dependency checkers | | Maintainability | SonarQube, CodeClimate, custom analysis | | Deployment | CI/CD metrics, deployment tracking |

5.2 CI/CD Integration¶

Actions: 1. Add fitness checks to CI pipeline 2. Configure gate conditions 3. Set up failure notifications 4. Create bypass procedures (with approval) 5. Document pipeline integration

Pipeline Integration Pattern:

# CI Pipeline
stages:
  - build
  - test
  - fitness_check  # ← Fitness function gate
  - deploy

fitness_check:
  script:
    - run-complexity-check --max 15
    - run-coverage-check --min 80
    - run-security-scan --fail-on critical
    - run-dependency-audit --max-age 90
  allow_failure: false

5.3 Dashboard Creation¶

Actions: 1. Create unified fitness dashboard 2. Display current vs target for each function 3. Show trend over time 4. Configure alerting rules 5. Set up regular reporting

Dashboard Components: | Component | Purpose | |-----------|---------| | Current Value | Real-time or latest measurement | | Target | Target value with threshold bands | | Trend | Historical trend (30/90/365 days) | | Status | Green/Yellow/Red indicator | | Last Updated | Measurement freshness |

Output¶

Measurement infrastructure
CI/CD integration
Fitness dashboards
Alert configuration

Step 6: Monitor & Report¶

Purpose¶

Continuously track fitness and communicate status.

Activities¶

6.1 Continuous Monitoring¶

Actions: 1. Monitor dashboard regularly 2. Respond to alerts promptly 3. Investigate degradation trends 4. Track improvement progress 5. Document significant events

Monitoring Cadence: | Type | Frequency | Action | |------|-----------|--------| | Automated alerts | Real-time | Respond per severity | | Dashboard review | Daily | Note trends | | Trend analysis | Weekly | Identify patterns | | Stakeholder report | Monthly | Communicate status |

6.2 Fitness Reporting¶

Actions: 1. Generate regular fitness reports 2. Highlight changes from last period 3. Explain any threshold breaches 4. Show progress toward targets 5. Recommend actions

Report Structure:

# Fitness Report - [Month Year]

## Executive Summary
Overall fitness score: 78/100 (↑ 3 from last month)

## Function Status
| Function | Current | Target | Status | Trend |
|----------|---------|--------|--------|-------|
| Response P95 | 280ms | 200ms | ⚠️ | ↓ improving |
| Availability | 99.95% | 99.9% | ✅ | → stable |
| ...

## Notable Events
- [Date]: Database optimization deployed (↓ 50ms latency)
- [Date]: Security vulnerability patched (CVE-2024-xxx)

## Recommendations
1. Continue performance optimization
2. Address increasing error rate trend

6.3 Degradation Response¶

When fitness degrades:

Identify affected function(s)
Determine if change-related or organic
Assess business impact
Create remediation plan
Track to resolution

Response Flow:

Alert → Triage → Investigate → Remediate → Verify → Document

Output¶

Regular fitness reports
Alert response records
Remediation tracking

Step 7: Review & Adapt¶

Purpose¶

Evolve fitness functions as the system and context change.

Activities¶

7.1 Periodic Review¶

Actions: 1. Schedule quarterly fitness review 2. Assess function relevance 3. Review target appropriateness 4. Identify missing coverage 5. Retire obsolete functions

Review Questions: | Question | Action if Yes | |----------|---------------| | Has business context changed? | Revisit priorities | | Are targets consistently met? | Consider raising targets | | Are targets never approached? | Reassess feasibility | | Are there unmeasured risks? | Add new functions | | Is a function rarely useful? | Consider retiring |

7.2 Target Adjustment¶

Actions: 1. Review consistently met targets 2. Identify stretch opportunities 3. Review consistently missed targets 4. Adjust based on learnings 5. Document changes with rationale

Adjustment Guidelines: | Situation | Adjustment | |-----------|------------| | Target met for 3+ months | Raise target 10-20% | | Target missed consistently | Investigate root cause | | External benchmark improved | Align to new benchmark | | Business priority changed | Reprioritize functions |

7.3 Function Evolution¶

Actions: 1. Add functions for new concerns 2. Split overly complex functions 3. Combine redundant functions 4. Improve measurement accuracy 5. Update automation

Output¶

Updated fitness function specifications
Adjusted targets
Review documentation

Supporting Workflows¶

Fitness-Guided Change Workflow¶

Purpose: Use fitness functions to evaluate proposed changes.

1. Propose Change
   - Document proposed modification
   - Identify potentially affected fitness functions

2. Predict Impact
   - Estimate fitness impact (improve/neutral/degrade)
   - For degradation, assess acceptability

3. Implement with Monitoring
   - Deploy change (canary/staged if risky)
   - Monitor affected fitness functions closely

4. Evaluate Results
   - Compare actual vs predicted impact
   - Roll back if unacceptable degradation

5. Document Learning
   - Record actual fitness impact
   - Update prediction models

Migration Fitness Workflow¶

Purpose: Guide migrations using fitness functions.

1. Baseline Source System
   - Measure all relevant fitness functions
   - Document baseline conditions

2. Define Target Fitness
   - Set fitness targets for migrated state
   - Identify fitness functions that must not degrade
   - Identify functions expected to improve

3. Create Migration Phases
   - Sequence migration for fitness preservation
   - Plan fitness validation at each phase
   - Define rollback criteria

4. Execute with Fitness Gates
   - Validate fitness after each phase
   - Proceed only if fitness maintained
   - Address degradation before continuing

5. Verify Final State
   - Confirm all target fitness achieved
   - Document improvements realized
   - Baseline for ongoing monitoring

Emergency Fitness Degradation Workflow¶

Purpose: Rapid response to critical fitness breaches.

1. Alert Received (Critical threshold breached)

2. Immediate Assessment (< 15 minutes)
   - Confirm measurement accuracy
   - Assess business impact
   - Identify recent changes

3. Triage Decision
   - Rollback recent change? → Execute rollback
   - Apply known fix? → Execute fix
   - Needs investigation? → Continue analysis

4. Mitigation (< 1 hour for critical)
   - Implement temporary mitigation
   - Communicate status to stakeholders
   - Monitor for improvement

5. Root Cause Analysis (< 24 hours)
   - Identify underlying cause
   - Plan permanent fix
   - Update fitness monitoring if needed

6. Post-Incident Review
   - Document incident timeline
   - Identify prevention measures
   - Update runbooks

Workflow Governance¶

Review Cadence¶

Activity	Frequency	Participants
Dashboard check	Daily	On-call engineer
Trend review	Weekly	Tech lead
Fitness report	Monthly	Architecture team
Full review	Quarterly	All stakeholders

Escalation Path¶

Severity	Response Time	Escalation
Critical	15 minutes	On-call → Team lead → Engineering manager
Warning	4 hours	Team lead → Architecture review
Trend degradation	1 week	Architecture team review

Success Metrics¶

Metric	Target
Fitness functions defined	100% of priority attributes
Measurement automation	> 80% automated
Alert response time	Within SLA
Target achievement rate	> 70% on track