The East African Cloud Tax
Running cloud infrastructure in East Africa means paying a hidden premium that most pricing calculators don't surface: data transfer costs to European and US regions are significantly higher than intra-region transfers, and many workloads route through distant availability zones by default.
Where the Money Goes
In our audits of new client infrastructure, we consistently find the same patterns:
- Oversized compute: Dev and staging environments running 24/7 on production-class instances
- Unoptimised storage: Hot S3/GCS storage for cold or archival data
- Data transfer waste: Application logs and backups routing to distant regions
- Idle resources: EIPs, NAT Gateways, and snapshots attached to terminated instances
Our Optimisation Playbook
1. Right-size compute using CloudWatch/Cloud Monitoring actuals, not estimated peaks 2. Implement auto-scaling with aggressive scale-down policies for non-production 3. Move cold data to Glacier/Archive storage tiers 4. Deploy a regional cache (CloudFront or Cloudflare) to reduce origin requests 5. Audit and delete unattached volumes, snapshots, and unused Elastic IPs monthly
Realistic Expectations
For a typical SME or NGO workload, we target 40–60% cost reduction in the first 90 days. Sustained savings require ongoing governance we recommend monthly cost review as part of any DevOps retainer.