Airbnb Optimizes Usage and Costs by Using Savings Plans and Actionable Cost Data on AWS

Airbnb’s journey to facilitating connection and community through travel began in 2007 when its first two hosts—and founders—welcomed three guests to their San Francisco home. Since then, the company’s hosts have welcomed over one billion total guest arrivals, with listings across more than 220 countries and regions.

From its early days as a company, Airbnb has powered its business using cloud services from AWS—a strategy that supports its rapidly expanding workloads as well as its need for agile compute and storage solutions. The company decided to invest in transforming its data and usage strategy on AWS. Because Airbnb operations involve hundreds of services and an agile technical culture, the company needed a strategic, centralized, and relatively frictionless approach to leaner purchasing.

Early in Airbnb’s journey, the company used a cost allocation visualization engine, AWS Cost Explorer, which has an easy-to-use interface that helps customers visualize, understand, and manage their AWS costs and usage over time. Later, to make use of its internal data infrastructure and extend the cost visualization capabilities of its data warehousing stack, Airbnb took advantage of data from the AWS Cost & Usage Report, which contains a comprehensive set of AWS cost and usage data and is available at an hourly, daily, or monthly level of granularity, as well as at the management or member account level.

Relying on its data warehouse, Airbnb uses the AWS Cost & Usage Report to run a custom pipeline that produces a dynamic view of Airbnb’s cost data and facilitates analytics. Airbnb’s pipeline ingests cost and usage files from Amazon Simple Storage Service (Amazon S3), which offers industry-leading scalability, data availability, security, and performance. To process files, Airbnb uses Amazon EMR, a cloud big data environment for processing vast amounts of data using open-source tools, making it possible for developers to run big data applications and petabyte-scale data analytics faster. Airbnb uses its pipeline to discount and amortize the data and then further customize it to support analytics and produce actionable insights and metrics.

To save on its use of Amazon Elastic Compute Cloud (Amazon EC2), which provides secure, resizable compute capacity in the cloud, Airbnb became an early adopter of Savings Plans, a flexible pricing model offering lower prices compared to Amazon EC2 On-Demand Instances in exchange for a specific usage commitment. Airbnb migrated its Amazon EC2 Reserved Instances to Savings Plans and discontinued its On-Demand Instance usage, resulting in cost savings. “Using Savings Plans has been a significant improvement to Airbnb’s cloud management process. It’s helped reduce our operational workload while also driving meaningful cost savings for our business,” says Ari Siegel, senior finance manager at Airbnb.

Airbnb also manages and saves on storage costs using Amazon S3 Intelligent-Tiering (S3 Intelligent-Tiering), a cloud storage class that delivers automatic cost savings by moving objects between four access tiers when access patterns change. By using S3 Intelligent-Tiering to manage storage locations for objects used by Airbnb’s data warehouse, Airbnb reduced storage costs by approximately 27 percent. Additionally, for replicated data stored for backup, the company uses Amazon S3 Glacier storage classes—a range of storage classes that users can choose from based on the data access, resiliency, and cost requirements of their workloads. Airbnb relies on the AWS Cost & Usage Report for hourly insights into data storage spending on Amazon S3.

Airbnb also manages costs within its logging infrastructure using Amazon OpenSearch Service, which makes it easy to perform interactive log analytics, near-real-time application monitoring, website searches, and more. To help lower costs, Airbnb employed UltraWarm storage for Amazon OpenSearch Service, which provides a cost-effective way to store large amounts of read-only data on Amazon OpenSearch Service. The solution contributed to a 60 percent reduction in Airbnb’s Amazon OpenSearch Service costs.

Airbnb’s finance and technology teams employed a broad range of AWS resources to build cloud efficiency and optimize operations. To work toward data-driven purchasing, the teams also expanded their consultation and engagement using AWS. And by delivering process and visibility improvements, the company helped empower individuals and teams to adopt cost management goals.