Key results:

• 52% total savings on BigQuery: By combining Rabbit’s reservation planning with automated Autoscaler optimization, Lufthansa reduced their BigQuery costs by 52%.
• 30% savings through automation: Rabbit’s automated BigQuery Autoscaler dynamically adjusted capacity, cutting waste and lowering costs by 30%.
• 33% savings on Cloud Run: Rabbit’s intelligent rightsizing ensured Cloud Run services were perfectly scaled, delivering an additional 33% in savings.

About Lufthansa Group

Lufthansa Group is one of the world’s largest and most respected airlines, serving millions of passengers and cargo destinations across the globe. The Lufthansa Group has partnered with Google Cloud on several modernization initiatives, including predictive maintenance with Lufthansa Technik (Google Cloud Story).

With Rabbit, Lufthansa continues to advance its cloud-first strategy, demonstrating that operational excellence and financial efficiency can go hand in hand.

The Challenge: Over-Provisioning Without Clear Utilization Insights

When Lufthansa’s platform team migrated workloads to Google Cloud, reliability was the priority. Engineers provisioned Cloud Run services with extra CPU and memory to make sure nothing broke in production.

That safety buffer added up. Since Cloud Run bills by vCPU-second and GiB-second, every over-request translated into direct cost, even if unused.

BigQuery carried another challenge: all queries ran on On-Demand pricing. This flexible model works well for unpredictable workloads, but Lufthansa’s daily, repeating queries made it unnecessarily expensive.

And while Committed Use Discounts (1–3 years) promised big savings, the team lacked clear visibility into usage patterns. Committing without that insight meant financial risk.

The consequence? A cloud bill that looked predictable in uptime terms, but spiraled higher than necessary every month.

Flying High, Spending Low at Google Summit 2025

At the 2025 Google Summit in Zurich, the Rabbit team and Lufthansa Group co-hosted a session showcasing how powerful and efficient the combination of BigQuery and Rabbit can be.

Watch the full session below to see how Lufthansa optimized performance and cost efficiency in BigQuery:

Utilizing BigQuery Reservations

The Lufthansa Group team faced challenges navigating BigQuery’s complex pricing model, particularly when deciding whether to stay on on-demand pricing or switch to reservations. With multiple workloads and varying usage patterns, it was difficult to identify where and how a reservation model would deliver the most value.

Rabbit’s Reservation Planner provided a solution. It automatically generated multiple scenarios across GCP projects to determine the most cost-effective pricing strategy. In addition, Rabbit offered intelligent recommendations for baseline and maximum slot configurations, backed by detailed usage metrics such as AVG, P90, P95, P99, and MAX slot utilization over recent periods.

After implementing Rabbit’s recommendations, Lufthansa achieved an impressive 52% reduction in BigQuery costs.

Automatically Reducing BigQuery Autoscaler Waste

BigQuery’s slot Autoscaler is a powerful feature that can scale computational resources within seconds after a large query is triggered. This flexibility helps meet demanding performance needs. However, there’s an important limitation: customers are billed for a minimum of 60 seconds for any provisioned slots, regardless of actual usage. As a result, organizations often face 50–60% resource waste, paying for far more capacity than their queries truly require.

Rabbit introduced an industry-first automation feature designed to minimize this waste. By dynamically adjusting the maximum slot setting in real time, updating every few seconds, Rabbit ensures that critical pipelines always have the power they need while idle or low-activity periods are handled cost-efficiently.

Once Rabbit Automation was activated, Lufthansa achieved an additional 30% cost reduction. Combined with previous optimizations, Rabbit helped the Lufthansa Group cut its overall GCP costs by nearly half.

See how much BigQuery waste your team could cut.

Calculate your waste →

Optimizing Cloud Run Service Utilization

When engineers deploy new services, it’s often difficult to precisely estimate the required CPU and memory resources. To ensure reliability, they typically choose higher configurations that guarantee performance but often lead to unnecessary infrastructure costs.

With Rabbit, teams can customize utilization recommendation settings based on the criticality of each application. For example, critical workloads can be configured to use maximum usage with an added 50% safety buffer, while less critical services can rely on average usage with a smaller buffer, such as 10%. This approach provides tailored, risk-aware recommendations that balance performance and efficiency.

For Lufthansa Group’s critical applications, the settings were tuned to use maximum usage plus a 100% buffer. Rabbit then automatically recalculated the optimal CPU and memory requests for each Cloud Run service. After applying these recommendations, Lufthansa achieved a 33% reduction in Cloud Run costs.

Summary

By leveraging Rabbit’s intelligent automation and optimization capabilities, Lufthansa Group significantly reduced costs across its Google Cloud environment, achieving 52% savings on BigQuery costs and 33% on Cloud Run.

Rabbit empowered Lufthansa to maintain top performance while operating more efficiently, proving that smart cloud management pays off.

Ready to see similar results?
Try Rabbit free for 30 days and start optimizing your Google Cloud costs today.

Start your free trial →