GAMING

Liquid Computing announced today that it has partnered with Dalhousie University’s Faculty of Computer Science to advance parallel On-line Analytical Processing (OLAP) on Very Large Databases (VLDB). The partnership project, which began in January of this year, has already demonstrated the dramatic performance advantages of applying new computing architectures to data warehousing problems. OLAP has become one of the most powerful and prominent technologies for knowledge discovery in VLDB environments. The size of contemporary data warehousing repositories has grown to a point where parallel processing solutions are mandatory for efficient and timely OLAP analysis. Dalhousie and Liquid Computing are partnering to research new parallel OLAP data cube algorithms executing on the fabric based LiquidIQ system. “We chose to partner with Liquid Computing on this project because it offers an opportunity to work with a unique next-generation computer architecture that offers new paths to improve the performance of data warehousing algorithms,” said Andrew Rau-Chaplin, Parallel Data Mining and OLAP Project Lead and Professor of Computer Science at Dalhousie. “Data warehousing applications and Very Large Databases are used extensively in industries with large data storage and analysis needs. We are seeing particular demand in location aware and spatial applications and from companies processing non-standard data for applications such as DNA and protein sequencing analysis. Companies are frequently challenged to rapidly analyze new data sets due to the computational barriers associated with VLDBs.” The underlying LiquidIQ fabric computing architecture converges computing, storage, networking, and broadband technologies that deliver industry leading performance and new functionalities that reduce operations costs. “We are using LiquidIQ to remove the common bottlenecks encountered with traditional compute clusters,” continued Rau-Chaplin. “LiquidIQ lets us remove the bottlenecks and still ride the cost curve of commodity processors, memory and IO components. We are seeing incredible performance from the LiquidIQ system and initial results indicate a two to three times improvement in performance metrics.” “This project is providing us with firsthand experience implementing leading-edge data warehousing applications on LiquidIQ,” said Keith Millar, Vice President of Product Line Management at Liquid Computing. “We are excited to be working with some of Canada’s leading researchers in the field of data warehousing to further investigate the potential for LiquidIQ and our customers.”