Contents lists available at SciVerse ScienceDirect
Information Systems
Information Systems 37 (2012) 391–392
0306-43
doi:10.1
journal homepage: www.elsevier.com/locate/infosys
Preface
Special Issue of DOLAP 2010 Information Systems
We welcome you to the special issue for the bestpapers presented at the 13th ACM Data Warehousing andOLAP Workshop 2010 (DOLAP) in Toronto, Canada. DOLAPcontinues its tradition of being a high quality forumwhere both researchers and practitioners in data ware-housing and On-Line Analytical Processing (OLAP) sharetheir findings on theoretical foundations, current data-base design methods, new trends and practical experi-ences. In recent years, research in these areas haveaddressed many topics, ranging from database designand software methodology issues, which help designersto build effective and useful decision-support applica-tions, to physical-level and query processing issues, aim-ing at increasing the performance of these applications inorder to process large amounts of data. However, thesuccessful use of data warehousing and OLAP technolo-gies within organizations has brought up new require-ments and research issues, in particular to cope with non-traditional application domains, such as text, biological,imaging, and spatio-temporal applications.
The DOLAP 2010 call for papers attracted 27 submis-sions. After careful review and discussion, the programcommittee accepted 8 full papers and 6 short papers. Outof the 8 full papers, we selected 4 papers to be invited forthe special issue in the Information Systems journal andafter a second round of reviews we finally accepted 3papers. Thus the relative acceptance rate for the papersincluded in this special issue is a competitive 10%. Need-less to say, these 3 papers represent innovative and high-quality research. We congratulate the authors of these 3papers and thank all authors who submitted articlesto DOLAP.
In general, research papers presented at DOLAP 2010show data warehousing is now a mature and wellestablished computer science field, but still with manyimportant open research questions. Relational databasesystems for data warehousing are the dominant technol-ogy: they are mature, reliable and efficient, but thecurrent growth of databases and the Internet are pushingthem to their limits. In addition, database modeling andsoftware engineering aspects remain challenging given
79/$ - see front matter & 2012 Published by Elsevier Ltd.
016/j.is.2012.02.001
the ever-changing nature of databases, both in structureand in size. New applications (outside traditional busi-ness processing), the integration with document retrieval(semi-structured data), faster hardware (multicore CPUs,flash memory, main memory database systems) and theInternet (data exchange, security) are making data ware-housing technology evolve even further and face newchallenges. Thus we expect research on data warehousingand OLAP to remain active in the future.
The paper titled ‘‘OLAP Query Reformulation in Peer-to-Peer Data Warehousing’’ by Golfarelli et al. advancesdistributed data warehouses. The paper proposes a peer-to-peer data warehousing architecture based on a net-work of heterogeneous peers, each equipped with anindependent data warehouse system that exposes queryanswering functionalities aimed at sharing business infor-mation in a peer-to-peer (P2P) distributed environment.To enhance the decision making process, an OLAP queryexpressed on a given peer is properly reformulated on theother peers. This paper presents an expressive languagefor the definition of mappings between the multidimen-sional schemata of peers and it also introduces a queryreformulation framework that relies on the translation ofthese mappings to relational schemata.
The paper titled ‘‘Incremental Maintenance of Materi-alized Views with Outerjoins’’ by Nica represents a bigstep forward in processing outerjoin queries, a well-known and difficult problem. This paper introduces sev-eral novel algorithms for the incremental maintenance ofmaterialized views with outerjoins (i.e. joins that includenon-matching rows, populating them with null values).The proposed algorithms relax the requirement for theexistence of the primary key attributes in the select list ofan outerjoin view to only a subset of the referencedtables. More important, the update algorithms do notrequire the null-intolerant property, except for somepredicates used in the view definition. Finally, the effi-cient maintenance of outerjoin views is implemented byusing exactly one update statement per view (supple-mented by merge substatements) for each table refer-enced in the view.
Preface / Information Systems 37 (2012) 391–392392
The paper titled ‘‘Time-HOBI: Index for OptimizingStar Queries’’ by Morzy et al. advances the state of theart in indexing for star join queries in data warehousing.The authors propose a novel bitmap index for optimizingstar joins and computing aggregations along dimensionhierarchies. The index is created on a chosen dimensionhierarchy and it has two main components: (1) a hier-archically organized bitmap index, having one bitmapindex per dimension level; (2) a time subindex thatimplicitly encodes time in every dimension.
Carlos Ordonez n
Department of Computer Science, University of Houston,
TX, USA
E-mail address: [email protected]
Il-Yeol SongCollege of Information Science and Technology,
Drexel University, PA, USA
E-mail address: [email protected]
n Corresponding author. Tel.: þ1 713 743 3350.
