Recursive queries were introduced as part of ANSI SQL 99 to support processing of hierarchical data typical of air flight schedules, bill-of-materials, data cube dimension hierarchies, and ancestor-descendant information (e.g. XML data stored in relations). Recently, recursive queries have also found extensive use in web data analysis such as social network and click stream data. Teradata implemented recursive queries in V2R6 using static plans whereby a query is executed in multiple iterations, each iteration corresponding to one level of the recursion. Such a static planning strategy may not be optimal since the demographics of intermediate results from recursive iterations often vary to a great extent. Gathering feedback at each iteration could address this problem by providing size estimates to the optimizer which, in turn, can produce an execution plan for the next iteration. However, such a full feedback scheme suffers from lack of pipelining and the inability to exploit global optimizations across the different recursion iterations. In this paper, we propose adaptive optimization techniques that avoid the issues with static as well as full feedback optimization approaches. Our approach employs a mix of multi-iteration pre-planning and dynamic feedback techniques which are generally applicable to any recursive query implementation in an RDBMS. We also validated the effectiveness of our proposed techniques by conducting experiments on a prototype implementation using a real-life social network data from the FriendFeed online blogging service.

Query processing in a DBMS typically involves two distinct phases: compilation, which generates the best plan and its corresponding execution steps, and execution, which evaluates these steps against database objects. For some queries, considerable resource savings can be achieved by skipping the compilation phase when the same query was previously submitted and its plan was already cached. In a number of important applications the same query, called a Parameterized Query (PQ), is repeatedly submitted in the same basic form but with different parameter values. PQ's are extensively used in both data update (e.g. batch update programs) and data access queries. There are tradeoffs associated with caching and re-using query plans such as space utilization and maintenance cost. Besides, pre-compiled plans may be suboptimal for a particular execution due to various reasons including data skew and inability to exploit value-based query transformation like materialized view rewrite and unsatisfiable predicate elimination. We address these tradeoffs by distinguishing two types of plans for PQ's: generic and specific plans. Generic plans are pre-compiled plans that are independent of the actual parameter values. Prior to execution, parameter values are plugged in to generic plans. In specific plans, parameter values are plugged prior to the compilation phase. This paper provides a practical framework for dynamically deciding between specific and generic plans for PQ's based on a mix of rule and cost based heuristics which are implemented in the Teradata 12.0 DBMS.

Relationship pattern based queries are important components of intelligence link analysis, Typically the analyst gives a prototypical graph pattern which needs to be approximately matched to the data. Performing such inexact graph pattern matching is currently carried out using some variant of graph edit distance measures. This approach suffers from two main shortcomings: (1) it relies on detailed graph edit cost assignment by the analyst, and (2) it cannot efficiently incorporate semantic similarities that can be, in most cases, computed based on appropriate ontologies. In this paper, we propose novel techniques for evaluating graph pattern queries to produce semantically-ranked results. Our approach systematically combines both partial structural matches and semantic similarities in order to relieve the user from specifying edit costs.

In numerous applications that deal with similarity search, a user may not have an exact idea of his information need and/or may not be able to construct a query that exactly captures his notion of similarity. A promising approach to mitigate this problem is to enable the user to submit a rough approximation of the desired query and use the feedback on the relevance of the retrieved objects to refine the query. In this paper, we explore such a refinement strategy for a general class of SQL similarity queries. Our approach casts the refinement problem as that of learning concepts using examples. This is achieved by viewing the tuples on which a user provides feedback as a labeled training set for a learner. Under this setup, SQL query refinement consists of two learning tasks, namely learning the structure of the SQL query and learning the relative importance of the query components. The paper develops appropriate machine learning approaches suitable for these two learning tasks. The primary contribution of the paper is a general refinement framework that decides when each learner is invoked in order to quickly learn the user query. Experimental analyses over many real life datasets and queries show that our strategy outperforms the existing approaches significantly in terms of retrieval accuracy and query simplicity.

Multi-relational data mining (MRDM) is concerned with data that contains heterogeneous and semantically rich relationships among various entity types. In this paper, we introduce multi-relational iceberg-cubes (MRI-Cubes) as a scalable approach to efficiently compute data cubes (aggregations) over multiple database relations and, in particular, as mechanisms to compute frequent multi-relational patterns ("item sets"). We also present a summary of performance results of our algorithm.