masthead-resources

Managing Transactions and Analytics Simultaneously in the High-Speed World of Financial Services

InterSystems IRIS Data Platform for High-Performance Transaction Management with Analytics - White PaperExecutive Summary
Successful financial services organizations today must be able to simultaneously process transactional and analytic workloads at high scale, processing billions of messages per day while supporting thousands of analytic queries per second from hundreds of applications, without incident. The consequences of dropped trades – or worse, a system failure – can be severe, incurring financial losses and reputational damage to the firm.

Traditional operational databases are often too slow to accommodate the high throughput and data-access rates required. These databases insert and update records in milliseconds, rather than delivering the microsecond response times that are needed, and are not optimized to concurrently support both transactional and analytic workloads at scale.

Today, in-memory databases are being used throughout the financial services industry, primarily due to their ability to support high-performance data-insert operations and analytic workload processing. However, in-memory databases alone have proved to be less than ideal, especially at handling mixed workloads as transaction volumes increase.

This white paper describes the requirements of transaction management and analytics platforms that must operate at very high scale without performance or reliability issues, even during periods of market volatility. It describes the strengths and weaknesses of the various technologies that have been used to date, and presents a new platform, InterSystems IRIS Data Platform. This platform is optimized for multi-workload processing at scale, delivering performance equal to or greater than in-memory technologies with the persistence and reliability of a traditional operational database.

For one leading global bank, InterSystems technology improved throughput by up to 500 percent and decreased latency by 1,000 percent compared with its previous in-memory DBMS-based system, and it has operated without incident since its implementation.

Download the White Paper
InterSystems IRIS Data Platform for High-Performance Transaction Management with Analytics