industrySolutions
information optimization

Next Generation Analytics

Tick Databases

Visualize Data Stored in Column-Oriented Tick Databases

Datawatch includes native connectors for SAP HANA, Sybase IQ, Kx kdb+, OneTick, and Thomson Reuters Velocity Analytics. Datawatch includes native connectors for SAP HANA, Sybase IQ, Kx kdb+, OneTick, and Thomson Reuters Velocity Analytics. Their extremely low processing latency makes column-oriented databases an excellent fit with Datawatch's real-time visualization tools.Column-oriented databases are excellent respositories for high volumes of tick data that must be retrieved quickly. These tick databases support higher information density than more common row-oriented relational databases. They are significantly more efficient in pulling data off disk and/or in caching it in memory.

Sort/Merge/Joins are also enormously faster since the column-tables are already sorted by Row ID, thus eliminating the need to sort before the merge. Many column-oriented databases are more than 100 times faster than similar row-oriented systems in terms of join performance.

Datawatch supports all major tick databases

Our data visualization tools support all major tick databases, including:

  • Kx kdb+
  • OneTick
  • SAP HANA
  • SAP Sybase IQ
  • Thomson Reuters Velocity Analytics

Perfect for tick data

Column-oriented databases are the preferred method for storing time series data in many applications in capital markets. For example, they are excellent choices for storing tick data from stock exchanges. Their speed and flexibility make them an excellent match with Datawatch's interactive dashboard tools for analyzing portfolio performance, risk analysis, equity screening, and similar financial applications.

In a typical financial services deployment, one or more column-oriented databases store tick data. These databases are then either connected directly to a Datawatch implementation or to a proprietary enterprise application that has Datawatch visual analytics tools embedded in it. In most cases, data from the column-oriented DBMS is combined through calculations with data from a standard row-oriented database, one or more real-time streaming feeds, and/or data from CEP engines. The resulting data is then displayed in specialized data visualizations optimized for historical analysis of time series data. This makes it easy to back-test trading algorithms, create faster, less costly compliance applications, view trade & quote data over long time periods, and analyze very large numbers of financial instruments, markets & geographies.