20 September 2011
Transtec is now offering ParStream, a highly performing database solution for big data analysis. It is faster by a factor of 1,000 than conventional systems while at the same time reducing the hardware capacity requirements by a factor of 20. ParStream has been developed by the start-up company of the same name based in Cologne, Germany.
ParStream makes efficient use of the high degree of parallelism of current multi-processor architecture - ranging from standard CPUs to highly parallel Nvidia Tesla GPUs. It also uses a novel indexing method, the Highly Parallel Compressed Index. Furthermore, ParStream can also execute queries on data immediately after a continuously running import, offering near real-time results, also with petabytes of data.
Application scenarios for ParStream exist at companies in a wide range of industries where big data analysis constitutes a challenge. Examples include network surveillance in the telecommunications industry, web analytics in e-commerce, smart grid management in the energy sector, or the misuse of credit cards in the financial sector. ParStream is also suitable for use in research fields such as climate model analysis or genetic research.