Scout24 Reshapes Global Hospitality Management with Geo-Partitioning and High Availability

For a quarter of a century, Scout24 has established itself as a central player in the digital real estate marketplace. Over 19 million monthly users, including homeowners, real estate agents, tenants, and buyers, depend on the Scout24 ecosystem to navigate the ins and outs of complex property transactions. The company has expanded its suite of offerings with the addition of the “Kotao” (for the hotel and restaurant management market) to serve the growing demands of its expanding customer base.

Kotao, an entity within the Scout24 group, delivers an all-in-one management system for hospitality and restaurants that helps hoteliers streamline their operations and enhance guest satisfaction. With globally positioned data centers for the hospitality industry, their system guarantees lightning-fast speeds and exceptional uptime, delivering instant responses for seamless guest service and eliminating slow load times.

• Streamlined property management for rooms and bookings
• Integrated Point of Sale (POS) systems for quick, easy, and secure transactions
• Unified system for room and table reservations with a 360° guest view

Technical Requirements for This Geo-Partitioned, Highly Available App

Like most of their services, Kotao is a modern application with complex requirements. The underlying database is critical, given how much the application depends on fast, accurate data.

The following parameters and requirements guide Kotao’s choice of an ideal database architecture:

• Deployment: Multi-region deployment across the EU, Asia Pacific, and the US.
• Primary Database Requirement: Built-in geo-partitioning and high availability. The database had to:
  • Support operations across multiple countries, including Germany, Spain, France, the UAE, Thailand, and Singapore. Such complexity necessitated a robust and globally distributed database with advanced geo-partitioning to meet different regional compliance requirements and data residency laws.
  • Ensure high availability and exceptional resilience, spanning multiple cloud regions and availability zones, to provide a seamless user experience. This requirement was a significant challenge with their previous AWS Aurora implementation.
• Cloud Used: AWS
• Cluster size (# of nodes): 72 nodes
• Workload characteristics: Read heavy, write heavy, 50/50 mix
• Replication factor: 3
• API Interfaces: YSQL (PostgreSQL)
• Additional Database Requirements:
  • PostgreSQL compatibility: Strong compatibility with PostgreSQL is crucial for smooth integration with existing applications, especially since their API services are written in Rust, which requires PostgreSQL driver compatibility.
  • ACID Compliance: The system must handle ACID transactions efficiently, which is critical for the financial payment systems within the application.
  • Horizontal Scalability: Since Kotao caters to different facets of hospitality management, from bookings to point-of-sale transactions, the database must be able to scale according to the demand without compromising performance.

Evaluating Databases for High Availability, Geo-Distribution, and More.

“In our quest for the perfect database solution, we thoroughly evaluated other players in the market like CockroachDB, ScyllaDB, MongoDB, Planetscale, and SurrealDB. Though YugabyteDB Managed stood out with its unparalleled Postgres compatibility and distributed architecture, ticking all our boxes. This was particularly crucial for us in handling ACID transactions within our financial payment systems.” said Nico Miebach, Scout24.

Scout24 transitioned from AWS Aurora to YugabyteDB Managed for their Kotao app to enable global reach with a multi-region deployment while also increasing the app’s resilience and availability.

Scout24 selected YugabyteDB Managed over Aurora, CockroachDB, ScyllaDB, and MongoDB because of the following key features:

• Rust/Postgres Compatibility:  YugabyteDB Managed’s advanced PostgreSQL compatibility was essential, especially considering Kotao’s API services were primarily written in Rust. The ability to switch out the credentials in the existing PostgreSQL driver without additional engineering work was a feature that most of the other databases lacked.
• Distributed Architecture: Kotao required a robust, globally distributed database that could handle the high demands of their platform, including unpredictable bookings traffic and point-of-sale transactions. YugabyteDB Managed’s distributed architecture delivers low latency and built-in resilience across multiple cloud regions and availability zones, which is a notable improvement over AWS  Aurora.
• Multi-Model Support and Strong Consistency: Kotao needed a database to support complex transactions with absolute consistency, particularly for their integrated payment platform and POS solution where transactional integrity is non-negotiable. YugabyteDB Managed’s multi-model support and strong consistency for ACID transactions were critical.
• Advanced Geo-Distribution: With the advanced geo-distribution options of YugabyteDB Managed, the Kotao app could meet compliance requirements while ensuring operational efficiency across multiple regions. This level of geo-distributed data management was essential for their planned expansion to 72 nodes across the EU, Asia Pacific, and the US.
• Scalability and Performance: YugabyteDB Managed’s architecture scales on demand, achieving targeted throughput for both read-intensive and write-intensive workloads. This scalable balance is crucial for hotel management applications that demand a 50/50 throughput mix of read and write operations.

Building a Globally Distributed, Highly Available Hospitality Management Application

Managing an extensive platform like Kotao required a modern, distributed database. The proposed architecture (based on YugabyteDB Managed) for the lightning-fast hospitality management solution includes:

1. Geo-partitioning to meet compliance

   The proposed architecture for the Kotao application is designed to use row-level geo-partitioning to manage data residency—in line with regulatory requirements like GDPR—and to lower latencies by storing data close to users. The architecture plans to designate a specific column in the database to dictate partitioning, with the data sorted into partitions that are then fixed to predetermined geographic locations via configurable metadata.

   This framework ensures data replication across selected zones or regions to adhere to locality and performance standards while also allowing dynamic scalability and streamlined data management. It will accommodate evolving needs by facilitating the addition or removal of data partitions to accommodate evolving requirements.

   With this setup, Kotao intends to ensure transactional consistency and facilitate seamless schema schema modifications across multiple regions, thereby providing users with fast, accurate, and reliable data access.

   

2. Rust Compatibility

   The architecture with YugabyteDB Managed as the foundation ensures Rust compatibility via the Rust-Postgres driver by adhering to the PostgreSQL wire protocol, allowing seamless database operations. It requires SSL for secure connections and is compatible with Rust-Postgres driver settings.

   YugabyteDB Managed also supports ACID-compliant distributed transactions and horizontal scalability to meet a Rust application’s need for data integrity and performance. This setup supports robust financial systems and any other application requiring reliable, low-latency access across geo-distributed clusters.

3. High Availability and Fault Tolerance

   The architecture built upon YugabyteDB Managed offers uninterrupted database service, ensuring no data loss (RPO=0) and rapid recovery (RTO=~3s) during outages. With data replicated across fault domains and auto-healing capabilities, it ensures continuous availability and zero-downtime upgrades. Proven at scale, it’s engineered to handle multiple simultaneous failures (from nodes to entire regions), ensuring robust, maintenance-free operations for mission-critical applications.

In Summary…

Scout24 enhanced its Kotao platform using YugabyteDB Managed, reshaping global hospitality management with a state-of-the-art, geo-distributed database. This strategic move ensures lightning-fast, resilient operations across continents. Perfectly aligning with their Rust infrastructure, YugabyteDB Managed delivers the robustness and PostgreSQL compatibility needed to keep them at the forefront of the digital hospitality/real estate revolution.

Learn more about YugabyteDB Managed and the features that Scout24 is leveraging through the following resources:

• Attend a Thursday demo to see geo-partitioning in action
• Sign up for a full-featured, free trial of YugabyteDB Managed to give geo-partitioning a try
• Visit our Docs and blog space to discover how you can build a global application with YugabyteDB