Gresham Computing has demonstrated that its reconciliation solution Clareti Transaction Control (CTC), which uses Intel and GigaSpaces technology, can process 500,000 transactions per second using ‘in memory’ matching techniques and technology. The ability to manage extreme processing in-memory is a key component of the GigaSpaces XAP solution that Gresham used as a foundation for the transactional control solution.
The vendor claims that the benchmarked performance is a new industry standard, equivalent to 1.8bn transactions per hour. The volume was achieved by running the application in a pure data environment and compute GigaSpaces grid, with no database attached. The achievement, recorded in extensive testing with Intel and GigaSpaces, dwarfs the benchmark of 50,000 transactions per second CTC set earlier this year, when using a database for reading and writing the data. Real world applications in corporate treasuries or banks processing trade transactions will, of course, be different depending upon the bandwidth available and a host of other factors.
Nevertheless Neil Vernon, development director at Gresham Computing, said: “The high performance achieved during this most recent benchmark shows that CTC can be used in extremely high-volume, low-latency environments to give operational certainty to financial organisations and infrastructure providers [and clients] conducting billions of transactions per hour in fast-moving markets.”
“CTC’s flexibility allows it to be optimised to operate in two modes: pure in-grid (in-memory) matching where CTC is used as a real-time service for the most exacting trading environments,” explains Vernon. “A full database attached allows a throughput of 50,000 transactions per second with all transactions written to the database.”
The tests were conducted using the Intel Xeon Processor E7 family, at Intel’s Computing lab in Reading, UK. All tests were executed multiple times to ensure repeatability and to calculate an average result, say the vendors. Gresham used a test data set of 150m transactions and modelled the data on a two-sided intersystem equity trade reconciliation. The match profile consisted of many-to-many matches, many-to-one matches, one-to-one exceptions and a proportion of the transactions left unmatched as singleton records.
CTC was built using the GigaSpaces XAP elastic application platform as an integral part of its infrastructure. XAP is geared for big data, employing an in-memory data grid for fast processing speed, ‘share-nothing’ partitioning for reliability and consistency, and an event-driven architecture that enables real-time processing of large volume streams and processing scalability. With CTC, organisations can rapidly replace existing manual and semi-automatic internal controls with automated controls, secure in the knowledge that CTC can scale to cope with higher volumes in the case of future growth or consolidation.
“The ability to process massive amounts of data in real-time from multiple sources, , while also retaining transactionality, scalability, and high availability is the IT holy grail for many financial services enterprises,” said Nati Shalom, chief technology officer (CTO) of GigaSpaces Technologies. “GigaSpaces in-memory technology integrates seamlessly with big data products, such as the NoSQL datastores employed by Gresham's CTC, to create a powerful and high-speed platform. The transactions, real-time event processing and elasticity of both processing and data all come at a fraction of the cost compared to traditional approaches.”