TimesTen

Originally developed within Hewlett-Packard labs, TimesTen was established as an independent entity in 1996 by co-founders Marie-Anne Neimat and Kurt Shoens. The company specialized in real-time data management software centered around in-memory database technology. Its core product, an in-memory relational database, was engineered to deliver microsecond response times and high throughput for online transaction processing (OLTP) applications. This performance is achieved by storing the entire database in physical memory (RAM) during runtime, which simplifies data access algorithms and shortens the code path compared to traditional disk-based systems.

The business model centered on providing infrastructure software for performance-critical applications. Its target markets included industries where processing speed is paramount, such as telecommunications for real-time billing, financial services for stock trading, call centers, and airline operations. Clients included major corporations like Cisco Systems, JPMorgan Chase, Sprint, and Nokia. The company was profitable with around 90 employees when it was acquired by Oracle Corporation in June 2005. Post-acquisition, TimesTen was integrated into Oracle's product line, enhancing its capabilities with features like PL/SQL support and integration with Oracle Enterprise Manager.

Now known as Oracle TimesTen In-Memory Database, the product operates as a memory-optimized relational database. It can function as a standalone database of record or as a high-performance cache for a larger Oracle Database, a model known as TimesTen Cache. This caching capability is designed for applications needing faster access to a performance-critical subset of data stored in a main Oracle database, with automatic synchronization between the cache and the primary database. The database supports standard SQL and PL/SQL interfaces, ensuring compatibility with existing applications and developer skills. For high availability, it utilizes transactional replication to a standby database, ensuring no data loss and minimal downtime in case of a server failure. The database is persistent, meaning transaction logs and checkpoints are saved to disk for recovery.

Keywords: in-memory database, relational database management system, OLTP, real-time data management, high throughput, low latency, database caching, Oracle TimesTen, data persistence, SQL interface, transactional replication, application tier database, financial trading systems, telecommunications billing, real-time applications, enterprise software, high availability database, memory-optimized database, data synchronization, PL/SQL support