Project events
1/2011 PhD student Rasmus Resen Amossen successfully defends his PhD thesis
02/2010 Meeting with Apptus (Lund) on follow-up project.
1/2010 Paper entitled On vertical partitioning of relational OLTP databases using integer programming accepted at SMDB 2010.
12/2009 Paper entitled Finding Associations and Computing Similarity via Biased Pair Sampling presented at ICDM 2009.
07/2009 Paper entitled “Secondary Indexing in One Dimension: Beyond B-trees and Bitmap Indexes” presented at PODS 2009.
05/2009 Paper entitled “Faster Join-Projects and Sparse Matrix Multiplication” presented at ICDT 2009.
04/2008 Fourth meeting with Apptus (ITU).
04/2008 Presentation of project results at SQL Server Open World 2008.
02/2008 Third meeting with Apptus (Lund).
08/2007 Second meeting with Apptus (ITU).
06/2007 PhD student starts. Kickoff meeting with Apptus.
03/2007 Project launched.
Project goals
The goal of the SQERD project (pronounced “squared”) is to apply modern algorithmic techniques to problems arising in database management systems in connection with evaluation of queries. We aim to develop algorithms that excel in two orthogonal directions:
-
Are very scalable (work well with increasing amounts of data). -
Are reliable in the sense that there is a solid mathematical understanding of their performance on any kind of data set they might be used on.
Previously, these goals have been achieved only for the simplest types of queries. We are aiming at extending the range of queries that are algorithmically well understood.
Scalable Query Evaluation
for Reliable Databases (SQERD)
for Reliable Databases (SQERD)
Time: 03/2007-10/2010
Funding agency: FTP
Budget: 2 million DKK
Host institution:
External partner:
People:
Rasmus Pagh (head)
Rasmus Resen Amossen (PhD student)
