Lock-in is usually framed as a technical problem. It is not. It is a financial problem. When your workloads are trapped on a single vendor’s platform, that vendor controls your costs. You cannot negotiate from a position of strength when the cost of leaving exceeds the cost of staying. Every hyperscaler knows this, and their pricing models are designed accordingly.
This has always been true. But in March 2026, with the Strait of Hormuz closed, oil above $100 per barrel, and global energy markets in turmoil, the financial exposure of cloud lock-in is about to become very visible on quarterly earnings calls.
Lock-In Is a Pricing Strategy
The hyperscalers do not accidentally make it expensive to leave. Egress fees, proprietary services, reserved instance commitments, and training ecosystems are all mechanisms that increase switching costs. Each one, individually, seems reasonable. Together, they form a financial moat that keeps customers paying whatever the vendor decides to charge.
Egress Fees: The Hotel California of Cloud
AWS charges $0.09 per GB for data transfer out. Azure is comparable. Getting data into these platforms is free. Getting it out costs real money.
For an enterprise with 50TB of data, just extracting your own data costs $4,500. If you are running a SaaS platform serving customers, egress fees become a permanent line item that scales with your success. You pay every month, and you cannot optimize it away without rebuilding your architecture.
This is not a fee for a service. It is a fee for leaving – or for doing business with anyone outside the walled garden.
Proprietary Services: The Technical Trap
Lambda, DynamoDB, Cosmos DB, Aurora Serverless, BigQuery – these services have no portable equivalent. Once you build on them, you are not migrating. You are rewriting. For a mid-market company, rewriting a production application is a six-to-twelve month project with significant risk. Most companies never do it. The hyperscalers know this.
Every proprietary service you adopt increases your switching cost. After three years on a platform, most enterprises have accumulated enough proprietary dependencies that migration is effectively off the table. That is when the price increases start.
Reserved Instances: Locking In Yesterday’s Prices
Reserved instances offer discounts in exchange for one-to-three year commitments. The pitch is simple: commit to spend, and you pay less per unit. But there is a catch. You are committing to a specific pricing structure with a vendor you cannot leave. If that vendor raises on-demand prices (making your “discount” less valuable relative to alternatives), or if your needs change, you are stuck.
Reserved instances are a bet that your vendor’s pricing will remain competitive for the duration of the commitment. In a stable market, that bet usually pays off. In a market where energy costs are spiking and hardware prices are surging, it is a bet with real downside.
Certification Ecosystems: Organizational Dependency
AWS has over 12 certification tracks. Azure has a comparable number. When your entire engineering team is certified on one platform, that certification investment creates organizational inertia. Proposing a migration means asking your team to retrain, and asking leadership to write off the training investment. This is a real cost, and it tilts every build-vs-buy and stay-vs-migrate decision toward the incumbent.
The Price Increases Are Already Here
Lock-in matters most when prices go up. And prices are going up.
| Event | Impact |
|---|---|
| AWS GPU instance pricing (Jan 2026) | 15% increase on P and G instance families |
| Azure EU local-currency pricing (Apr 2023) | 6-11% increase for European customers |
| DDR5 memory spot pricing (Sep 2025 – present) | Up 307%, with Dell and Lenovo announcing 15-20% server price hikes |
| Strait of Hormuz closure (Mar 2026) | Oil above $100/barrel, US energy costs rising |
These are not isolated events. They form a supply chain: energy costs drive data center operating expenses, memory and hardware costs drive procurement expenses, and both flow through to cloud pricing. Analysts widely expect Q2 and Q3 2026 to bring additional cloud price adjustments as these procurement costs work through vendor billing cycles.
The Energy Cost Equation
Data centers globally are projected to consume 75.8 GW of power in 2026. The Strait of Hormuz handles roughly 20% of global oil transit. Its closure does not just raise fuel prices – it disrupts the entire energy supply chain, including natural gas markets that power a significant share of US electricity generation.
Hyperscalers pass energy costs through to customers, either directly via price increases or indirectly via reduced discount availability. If you are locked into a platform with no alternative, you absorb those increases with no leverage to negotiate. Your vendor’s energy cost problem becomes your budget problem.
What Lock-In Actually Costs You
The total cost of lock-in is not just the premium you pay on any single service. It is the compound effect of being unable to respond to market changes.
- You cannot renegotiate effectively. Your vendor knows your switching costs. So do their account managers. Every renewal conversation starts from that asymmetry.
- You cannot hedge. Multi-cloud strategies are theoretically possible but operationally difficult when your applications depend on proprietary services.
- You cannot budget accurately. Variable pricing plus vendor-controlled rate changes mean your cloud spend is a forecast, not a budget. Finance teams hate this.
- You absorb external shocks. Energy price spikes, hardware shortages, and geopolitical disruptions all flow through to your bill, and you have no mechanism to absorb them except paying more.
For a company spending $500,000 annually on cloud infrastructure, even a 10% price increase driven by factors outside your control represents $50,000 in unplanned spend. Over a three-year reserved instance commitment, that exposure compounds.
The Alternative: Infrastructure You Can Leave
The antidote to lock-in is portability. If your workloads run on standard APIs and open infrastructure, you can move. That ability to move is what gives you negotiating leverage, budget predictability, and protection against external cost shocks.
OpenStack provides this portability. It exposes standard compute, storage, and networking APIs that are not tied to any single vendor. Workloads built on OpenStack APIs can be migrated between providers using existing tools – Terraform, Ansible, standard CLI clients. There is no rewrite required, no proprietary service to replace, and no egress fee to escape.
Open Edge’s Approach
Open Edge Cloud is built on OpenStack and designed for enterprises that want cost predictability without sacrificing infrastructure quality.
- Contract-based fixed pricing. Your price is locked for the term of your contract. No mid-term adjustments, no energy surcharges, no surprise line items. When oil hits $100 per barrel, your cloud bill does not change.
- Zero egress fees. Move data in and out freely. Your data is your data, not a revenue stream for your cloud provider.
- No proprietary services. Standard OpenStack APIs for compute, block storage, object storage, networking, load balancing, and identity. Everything you build is portable.
- No energy pass-throughs. Energy cost volatility is our problem to manage, not yours. Fixed pricing means fixed pricing.
- FIPS 140-3 validated encryption (CMVP Certificate #5115) on all data at rest. Our platform follows SOC 2 and ISO 27001 control frameworks.
- US-sovereign infrastructure. All hardware is in US data centers (Iron Mountain VA-1, Manassas, Virginia). No multinational supply chain exposure.
Migration Is Simpler Than You Think
If your workloads run on virtual machines, block storage, and standard networking, migration to OpenStack-based infrastructure is straightforward. The APIs are well-documented, the tooling is mature, and the process is well-understood. Terraform providers, Ansible modules, and CLI tools all work with OpenStack without modification.
For workloads currently running on proprietary services, migration requires more planning – but the investment pays for itself in cost predictability and vendor independence. Every workload you move to standard APIs is a workload that can never be held hostage by a single vendor’s pricing decisions.
The Window Is Now
Cloud price increases do not arrive with advance notice. By the time your Q2 bill reflects the new reality, your options are limited. Renegotiating takes months. Migration takes longer. The time to evaluate alternatives is before the price increases hit, not after.
If your organization is spending six or seven figures annually on cloud infrastructure, and your current provider controls your ability to leave, you are carrying financial risk that does not show up on any balance sheet – until it does.
Lock-in is not a technical problem you can solve later. It is a financial exposure you are carrying right now.
Talk to Us
Open Edge Cloud provides enterprise cloud infrastructure with contract-based fixed pricing, zero egress fees, and no proprietary lock-in. If you are evaluating your options before the next round of price increases, our team can walk through your current environment and show you what a migration path looks like.