October 2007 Archives

Large databases require the use of parallel computing resources to get good performance. There are several fundamentally different parallel architectures in use today: shared memory, shared disk, and shared nothing. This post examines each approach in terms of how it impacts database scalability. Continue reading "Database parallelism choices greatly impact scalability" »

DeWitt and Madden discuss how trends in CPU technology, similar to trends in mass storage technology, will increasingly drive the use of column-store architecture in database systems designed primarily for processing decision support queries. They discuss the impact of slotted pages in row stores and how column stores avoid the CPU performance penalty. Continue reading "CPU trends, like disk trends, will favor adoption of column stores" »

Mike Stonebraker offers his personal highlights of VLDB 2007, including a historical note on RDBMS design, the need to improve database administration interfaces, new capabilities and tools from Yahoo!, and some thoughts on the OLPC project. Continue reading "Thoughts on the VLDB conference" »

Ten years after its publication, Mitch Cherniack looks at the relevancy and applicability of the "The Database Warehouse Toolkit." He discusses four issues that he contends reveal the deficiencies of traditional relational row-store OLTP engines when used for data warehouse applications. Continue reading "Reflections on the Kimball Data Warehouse "Bible": Time for a New Testament?" »