April 2018 Summaries

CockroachDB has released a whitepaper demonstrating its high OLTP performance, achieving over 128,000 tpmC on a TPC-C dataset of over 2 terabytes, which is over 10 times the throughput of Amazon Aurora in similar conditions. This performance was achieved in serializable isolation mode, offering both strong isolation and high performance without sacrificing correctness and maintainability. The benchmark, though unofficial and not validated by the Transaction Processing Council, uses a richer set of SQL semantics compared to NoSQL benchmarks. CockroachDB's linear scalability is highlighted, showing that a 10,000 warehouse setup using 30 nodes offers 10 times the throughput of a 1,000 warehouse setup with 3 nodes, demonstrating that operational costs can scale with business needs. The whitepaper adheres to CockroachDB's open-source philosophy by providing a step-by-step guide to reproduce these performance numbers, inviting scrutiny and contributions to ensure transparency and accuracy.

CockroachDB introduces geo-partitioning and archival-partitioning as features to enhance data management in distributed databases, aiming to improve performance and compliance. Geo-partitioning reduces latency by keeping data close to users' locations, aligning with regulations like GDPR by controlling data domiciling, and improving user experience through reduced data travel distances. Archival-partitioning optimizes costs by moving infrequently accessed data to slower, cheaper storage, which is particularly beneficial for compliance and cost efficiency. CockroachDB differentiates itself by automatically sharding data and using a SQL planner to optimize query performance, obviating the need for manual partitioning often required in other databases. These features, coupled with CockroachDB’s automatic replication and online schema changes, are designed to streamline database management without disrupting operations, and there are plans to enhance these capabilities further to support seamless compliance and data handling.

CockroachDB 2.0 is an advanced open-source database designed to enhance cloud applications by maintaining SQL capabilities while offering improved scalability, performance, and flexibility. Targeting companies facing distributed scale, resilience, and consistency issues, it supports rapid development through JSON data types, enabling swift prototyping and zero-downtime schema changes. The database's ability to scale seamlessly by adding nodes, rather than relying on monolithic relational databases, ensures robust performance even under significant growth, as evidenced by its superior throughput in industry benchmarks. Additionally, CockroachDB 2.0 caters to global user demands with features like multi-regional data capabilities, geo-partitioning, and advanced visualization tools, supporting applications' adaptability to changing data domiciling requirements while maintaining low-latency operations. This innovative release aims to simplify data management across private, public, and hybrid cloud environments, easing the burden of architecture planning and middleware management, thus allowing developers to focus on building impactful applications.