Know What You Have
You can't migrate what you don't know exists. You can't protect what you haven't mapped.
Do you have a current, accurate CMDB (Configuration Management Database)?
A CMDB tracks what IT assets you have, where they are, and how they relate. Most organizations have one—but it's often 6-12 months out of date.
How many IT assets does your organization manage?
This includes servers, VMs, SaaS applications, databases, network devices. A rough estimate is fine—this determines your discovery approach.
What percentage of your IT infrastructure runs on US-based cloud providers?
Include AWS, Azure, GCP, and SaaS platforms (Microsoft 365, Salesforce, etc.). This is your geopolitical exposure.
Have you mapped dependencies between your systems?
Do you know which systems depend on each other? If System A fails, which others stop working?
Implementation Guides
If You Have a Working CMDB
Extract these critical fields for each asset:
- Asset Name & Type (server, SaaS, application, database)
- Hosting Location (cloud provider, datacenter, on-premises)
- Geographic Region (US, EU, NL)
- Business Criticality (Tier 1/2/3)
- Dependencies (what it depends on, what depends on it)
- Data Classification (contains sensitive/personal/financial data?)
- Recovery Time Objective (max acceptable downtime)
- Owners (business owner + technical owner)
Export to Breakout Inventory Template
Create a spreadsheet with these columns:
Asset_Name, Type, Provider, Location, Tier, Dependencies, RTO_Hours, Business_Owner, Tech_Owner
Validation Check
Once exported, filter and validate:
- Filter: Provider = Microsoft/AWS/Google/US-based → This is your geopolitical exposure
- Sort by Tier 1 (mission-critical) → These must have breakout capability first
- Check for missing owners → No owner = no accountability during crisis
- Verify RTO values → Unrealistic RTOs doom your breakout plan
You Have 2 Weeks to Build This
Without a CMDB, you need rapid automated discovery. Here's the sprint plan:
Week 1: Automated Discovery
Cloud Assets (Azure)
# Login to Azure
Connect-AzAccount
# Export all resources
Get-AzResource | Select-Object Name, ResourceType, Location, ResourceGroupName, Tags |
Export-Csv -Path "azure-inventory.csv" -NoTypeInformation
# Filter production resources by tags
Get-AzResource -TagName "Environment" -TagValue "Production" |
Select-Object Name, ResourceType, Location |
Export-Csv -Path "azure-production.csv" -NoTypeInformation
# Get subscriptions overview
Get-AzSubscription | ForEach-Object {
Set-AzContext -Subscription $_.Id
Get-AzResource | Measure-Object
}
Cloud Assets (AWS)
# List all EC2 instances aws ec2 describe-instances --query 'Reservations[*].Instances[*].[InstanceId,InstanceType,State.Name,Tags[?Key==`Name`].Value|[0]]' --output table # Export S3 buckets aws s3api list-buckets --query 'Buckets[*].[Name,CreationDate]' --output table # List RDS databases aws rds describe-db-instances --query 'DBInstances[*].[DBInstanceIdentifier,Engine,DBInstanceStatus]' --output table
On-Premise Network Scan
# Basic network sweep (find active hosts)
nmap -sn 10.0.0.0/24 -oG network-scan.txt
# Service detection on discovered hosts
nmap -sV 10.0.0.1-254 -oX network-services.xml
# Deep scan with OS detection (WARNING: intrusive)
sudo nmap -A -T4 10.0.0.0/24 -oA deep-scan
# Parse results to CSV
grep "Host:" network-scan.txt | awk '{print $2","$3}' > active-hosts.csv
SaaS Inventory
SaaS is harder to discover automatically. Use these sources:
- Finance: Review credit card statements for recurring SaaS charges
- Procurement: Check vendor contracts and purchase orders
- SSO logs: Export users/applications from Azure AD, Okta, or your identity provider
- Browser extensions: Tools like Torii or Zylo can discover shadow IT
# Get all enterprise apps registered in Azure AD Connect-AzureAD Get-AzureADServicePrincipal -All $true | Select-Object DisplayName, AppId, PublisherName | Export-Csv -Path "azuread-apps.csv" -NoTypeInformation
Week 2: Business Validation
Automated discovery finds assets, but you need business context. For each discovered system, ask the business owner:
- What does this system do? (1 sentence description)
- Who uses it? (team, department, external customers)
- Can it be down for 24 hours? (yes/no)
- Does it contain critical data? (customer, financial, operational)
Send this email template to business owners:
Subject: [URGENT] IT Asset Validation - [System Name] Hi [Owner Name], We're conducting a risk assessment of our IT infrastructure. For [System Name], we need the following information: 1. Function: What does this system do? (1 sentence) 2. Users: Who depends on it? 3. Downtime tolerance: Can it be offline for: [ ] 1 hour [ ] 24 hours [ ] 1 week [ ] Longer 4. Data sensitivity: Does it contain: [ ] Customer data [ ] Financial data [ ] Operational data [ ] None Please respond within 48 hours. No response = we'll assume "not critical." Thanks, [Your Name] IT Risk Management
Consolidate Results
Merge automated discovery + business validation into a master inventory:
- Tab 1: Cloud Assets (Azure/AWS/GCP exports)
- Tab 2: SaaS Applications (from SSO/finance/procurement)
- Tab 3: On-Premise Systems (Nmap results + validation)
- Tab 4: Dependencies (see Infoblok C)
Why Dependencies Matter
System A might look EU-hosted and safe. But if System A depends on Database B in AWS us-east-1, you have transitive US exposure. Hidden dependencies kill breakout plans.
Step 1: Network Flow Analysis (Technical)
Capture actual traffic to see which systems communicate:
# Capture 10,000 packets on eth0 sudo tcpdump -i eth0 -c 10000 -w capture.pcap # Filter traffic to/from specific subnet sudo tcpdump -i eth0 net 10.0.0.0/24 -w subnet-capture.pcap # Analyze captured traffic tshark -r capture.pcap -T fields -e ip.src -e ip.dst -e tcp.dstport | sort | uniq -c | sort -rn > flow-analysis.txt # Output: "count source_ip dest_ip port" # Example: "523 10.0.1.15 10.0.2.30 3306" = 523 connections from .15 to .30 on MySQL port
Azure Network Watcher (for cloud)
# Enable NSG flow logs
$nsg = Get-AzNetworkSecurityGroup -Name "prod-nsg" -ResourceGroupName "prod-rg"
$storageAccount = Get-AzStorageAccount -ResourceGroupName "logging-rg" -Name "flowlogsstorage"
Set-AzNetworkWatcherFlowLog `
-NetworkSecurityGroup $nsg `
-StorageAccountId $storageAccount.Id `
-Enabled $true `
-RetentionInDays 7
# Query flow logs
Get-AzNetworkWatcherFlowLog | Where-Object {$_.Enabled -eq $true}
Step 2: Application Owner Interviews (Business)
Technical analysis finds connections. Business interviews explain why they exist and what breaks if they fail.
Ask application owners:
- Which databases does your application use?
- Which external APIs are called? (payment, shipping, authentication)
- Which other internal systems are required? (inventory, CRM, ERP)
- What happens if [dependency X] is unavailable for 1 hour? 1 day?
- Are there any undocumented integrations? (scripts, cron jobs, manual processes)
Document answers in a structured format:
Application: [Name] Owner: [Name] Interview Date: [Date] DEPENDENCIES: 1. Database: [Name, Location, Critical Y/N] 2. APIs: [List with providers] 3. Internal Systems: [Names] 4. External Services: [SaaS, cloud services] FAILURE IMPACT: - If [Dependency X] fails: [Impact description] - Workarounds: [Manual processes, alternatives] - Maximum tolerable downtime: [Hours/Days] HIDDEN INTEGRATIONS: [Any scripts, cron jobs, or undocumented connections]
Step 3: Visualize Dependencies
Build a dependency graph to spot exposure chains:
digraph Dependencies {
rankdir=LR;
node [shape=box, style=filled];
// EU Systems (green)
AppA [fillcolor=lightgreen, label="App A\n(EU)"];
AppB [fillcolor=lightgreen, label="App B\n(EU)"];
// US Systems (red)
AzureAD [fillcolor=lightcoral, label="Azure AD\n(US)"];
Database [fillcolor=lightcoral, label="SQL DB\n(AWS US)"];
// Dependencies
AppA -> Database;
AppA -> AzureAD;
AppB -> AppA;
AppB -> Database;
}
// Generate image: dot -Tpng dependencies.dot -o dependencies.png
Or use online tools:
- Mermaid Live Editor: https://mermaid.live - web-based, no install needed
- draw.io: https://app.diagrams.net - drag-and-drop visual editor
- Lucidchart: Commercial option with templates
Step 4: Identify Transitive Exposure
Once you have the dependency graph:
- Color-code by location: EU = green, US = red
- Trace paths: Does any EU system depend on US systems?
- Flag critical paths: Tier 1 apps with US dependencies = highest priority for breakout
- Calculate exposure score: [Direct US systems] + [EU systems depending on US] = Total Exposure
Output: Dependency Matrix
Create a master table:
System_A, Depends_On, Location_A, Location_B, Critical, Transitive_US_Exposure App_CRM, Azure_AD, EU, US, Yes, Yes App_CRM, SQL_Database, EU, US, Yes, Yes App_Warehouse, App_CRM, EU, EU, No, Yes (via CRM->Azure) App_Finance, SQL_Database, EU, US, Yes, Yes
Use this matrix to:
- Prioritize which dependencies to break first (Tier 1 + US exposure)
- Identify alternatives (can we switch Azure AD to local LDAP?)
- Estimate breakout complexity (more dependencies = harder migration)
Calculate Your Risk
Once you have inventory and dependencies, calculate your geopolitical exposure score for each asset:
Scoring Formula
Per Asset Score = (Provider + Data + Lock-in + Alternatives) × Criticality
Provider Location (0-3 points)
- 0 pts: Netherlands-based (local datacenter, NL cloud)
- 1 pt: EU-based (not NL, but within EU jurisdiction)
- 3 pts: US-based (AWS, Azure, GCP, Microsoft 365, etc.)
Data Sovereignty (0-4 points)
- +0 pts: No sensitive data (test environments, logs)
- +2 pts: Personal data subject to GDPR
- +2 pts: Financial data (invoices, transactions, payroll)
- +2 pts: Operational data (if loss stops business)
Note: Scores are cumulative. An asset with personal + financial data = +4 pts.
Contract Lock-in (0-5 points)
- +0 pts: No contract or month-to-month
- +1 pt: 1-2 year contract, reasonable exit terms
- +2 pts: 3+ year contract
- +2 pts: Exit fees > €100k
- +1 pt: Data export difficult (no API, manual only)
Alternative Availability (0-3 points)
- +0 pts: Easy alternatives exist (commodity services)
- +1 pt: Alternatives exist but require migration effort
- +2 pts: Limited alternatives (specialized platforms)
- +3 pts: No viable alternatives (vendor-specific functionality)
Criticality Multiplier (×1 to ×3)
- ×3: Tier 1 (mission-critical, business stops without it)
- ×2: Tier 2 (important, significant impact if down)
- ×1: Tier 3 (nice-to-have, minimal impact)
Example Calculation
Asset: Microsoft 365 (Email + SharePoint)
- Provider: US-based = 3 pts
- Data: Personal data (emails) = +2 pts
- Lock-in: 3-year EA, difficult export = +3 pts
- Alternatives: Nextcloud/ownCloud exist = +1 pt
- Criticality: Tier 1 (email is mission-critical) = ×3
Total Score: (3+2+3+1) × 3 = 27 points (HIGH RISK)
Asset: Internal Wiki (on-prem)
- Provider: NL datacenter = 0 pts
- Data: No sensitive data = +0 pts
- Lock-in: Self-hosted, open source = +0 pts
- Alternatives: Many wiki platforms = +0 pts
- Criticality: Tier 3 (useful but not critical) = ×1
Total Score: (0+0+0+0) × 1 = 0 points (NO RISK)
Interpreting Scores
- 30+ points: Critical exposure. Must have breakout plan immediately.
- 15-29 points: High risk. Prioritize for Phase 2 planning.
- 5-14 points: Medium risk. Address after critical systems.
- 0-4 points: Low risk. Monitor but not urgent.
Build Your Prioritized List
Sort all assets by score (highest first). Your top 10-20 assets are your breakout priorities. These need alternative infrastructure first (Phase 2).
Rank, Asset, Score, Action 1, Microsoft 365, 27, Build Nextcloud alternative + email migration plan 2, Azure SQL Database, 24, Migrate to EU PostgreSQL cluster 3, Salesforce CRM, 21, Evaluate EU CRM alternatives or hybrid approach 4, AWS S3 Storage, 18, Replicate to EU object storage 5, GitHub Enterprise, 15, Deploy GitLab on-prem or EU region
