McObject Improves Performance of eXtremeDB Financial Edition and Eases Development with Version 6.0

Blog entry

McObject, a provider of low latency, high performance database management technologies, will release eXtremeDB Financial Edition version 6.0 today. The new version is a major upgrade including support for big data management for capital markets analytics and access to the database’s vector-based statistical functions using SQL and Python.

Version 6.0 of eXtremeDB Financial Edition has been tested by several McObject users and is now in production. Existing users of the software can upgrade under the company’s standard support and maintenance programme, while others looking for a high performance database with extensive functionality are expected to join them.

Upgrades to eXtremeDB Financial Edition that support big data processing include a distributed query processing capability that partitions a database across multiple servers and distributes query processing across multiple CPUs to achieve parallel computing that can deliver significantly improved processing performance.

The upgrade also adds the run-length encoding compression algorithm that reduces the size of stored market data, in turn reducing storage costs and accelerating processing. McObject says tests using the market data compression facility with Chicago Board Options Exchange Market Volatility Index tick data reduced the data to a quarter of its pre-compression size and increased the speed of reading the database by 21%.

eXtremeDB Financial Edition version 6.0 also extends the software’s library of vector-based statistical functions to about 200 functions that are pipelined in CPU cache to minimise latency when analysing market data. The company already offers access to the library using its own application programming interface, but in Version 6.0 adds access using the SQL and Python languages.

The addition of SQL, which can be used in applications written in languages including C++, Python, Java and C#, enables faster development and increases the pool of developers with skills to work with the database system. The inclusion of Python adds a language that supports fast deployment of tasks such as rapid prototyping. Python can also be used with eXtremeDB’s dynamic database definition language to implement ideas quickly and optimise them rapidly by testing changes to code, database tables and indexes.

Chris Mureen, chief operating officer at McObject, says uses of eXtremeDB Financial Edition span from algo trading to matching engines, order books and risk management, a sweet spot for the company’s technology. Looking forward, he expects eXtremeDB Financial Edition to be particularly suited to algo trading and risk management applications that often analyse terabytes of data. While most of McObject’s customers are based in the US, the company has a couple of users in India and China, and recently signed MCO Europe as its exclusive distributor in Europe.