University of Ottawa - Carleton University
Ottawa-Carleton Institute for Computer Science (OCICS) Presentation
|
February 15, 2013 @ 10:00a.m.
Parallel Real-time OLAP on Multi-core Processor
|
Speaker: Xiaoyun Zhou
Location: STE 5084 (EECS building conference room) |
ABSTRACT
Today, successful organizations increasingly rely on data analysis to develop new opportunities, guide business strategies and optimize resources. Online analytical processing (OLAP) is one of the most powerful technologies to provide user interactively analyze multidimensional data from multiple perspective. Traditionally, OLAP system loads data in batches periodically (e.g. weekly), then builds static data cubes to answer queries quickly. However, many businesses expect a real-time OLAP which updates itself immediately when new data arrives and allows performing analysis with integrated data source on the fly. A significant performance issue arises in such real-time OLAP with large scale data warehouses. In this presentation, a parallel real-time OLAP system for multi-core processors will be introduced. The system provides a dynamic data index structure to handle data insertions and range queries in parallel on multiple cores. Some experiments were performed on different infrastructures such as physical machines with different number of cores, Amazon Cloud instances, and the Ontario Research and Education VCL Cloud instances. Through the tests, we not only achieved the significant speedup but also observed how the parameters influenced the performance.
|
| Return to Schedule |
|
