I'd rather play golf

Yesterday afternoon I listened to Sunil Prabhakar of Purdue University on the Orion Uncertain Data Management System [1-4]. Sunil’s talk is one of the Database Seminar Series at the University of Waterloo which is sponsored by Sybase iAnywhere. Here’s Sunil’s abstract: Many applications domains are faced with the need to store and manipulate uncertain or [...]

[Read more →]

On Wednesday I attended a lecture entitled “Towards a Theory of Search Queries” presented by Dirk Van Gucht of Indiana University. The talk summarized the results that were published earlier this year in a paper [1] of the same name at the 12th International Conference on Database Theory (ICDT) (ACM version is here, freely available [...]

[Read more →]

At June’s 2009 ACM SIGMOD Conference, Surajit Chaudhuri of Microsoft Research presented a thought-provoking paper [1] entitled “Query Optimizers: Time to Rethink the Contract?”. Here is the abstract: Query Optimization is expected to produce good execution plans for complex queries while taking relatively small optimization time. Moreover, it is expected to pick the execution plans [...]

[Read more →]

This morning Chris Olston of Yahoo! Research presented a talk entitled Pig: High-Level Dataflow on top of Map-Reduce as part of the iAnywhere-sponsored Database Research Seminar series at the University of Waterloo’s School of Computer Science. Pig Latin [1] is a SQL-like language that I first heard described in a talk given by Utkarsh Srivastava [...]

[Read more →]

This afternoon I attended a talk at the University of Waterloo presented by Matei Zaharia, a PhD student at UC Berkeley (and a former UWaterloo undergraduate). Matei’s talk was entitled “Trends and Research Challenges in Cloud Computing” but primarily his talk centered around scheduling innovations for Hadoop, Apache’s implementation of Map-Reduce [1], using Matei’s internship [...]

[Read more →]

Oege de Moor of Semmle Ltd pointed out to me that the latest edition of SemmleCode for Java program analysis, entitled SemmleCode Professional, is now available as a private preview for registered users. I’m one of them and I’m looking forward to trying the new version out for a test drive–continuing with my running example [...]

[Read more →]

Bill-of-materials (transitive closure) queries are reasonably common. Without recursion, the only way to compute a bill-of-materials result with a single SQL statement is through self-outer-joins, and the problem with self-joins, of course, is that one has to know in advance how many levels to traverse–that is, how many outer joins to code within the query. [...]

[Read more →]

Over the next few weeks, I’ll be posting a series of articles about the differences between various object-relational mapping (ORM) packages, including .QL from Semmle, Hibernate and nHibernate from JBoss/Redhat, and LINQ from Microsoft. In a real sense I’ll be joining a chorus of individuals making such comparisons but much of this discourse has omitted [...]

[Read more →]

Apache’s Hadoop is getting a lot of traction amongst academics and commercial data warehouse vendors, and it may come to pass — relatively quickly — that aspects of Hadoop’s parallel-processing capabilities will be integrated with traditional relational database systems. That integration may occur with the help of the integration of Yahoo!’s Pig. In a recent [...]

[Read more →]

Going through the individual contributions to the 2008 Claremont database research self-assessment report makes interesting reading. For example, Joe Hellerstein wonders if declarative languages such as Datalog (or their ilk) can make a comeback: perhaps a tighter integration of application logic and database query capability (without SQL) is essential to improving programmer productivity. Oege de [...]

[Read more →]