Archive

Archive for the ‘Accepted Papers’ Category

SIGMOD 2012 Accepted Paper

April 1st, 2012
Comments Off

The Brown Data Management Research Group has the following paper in SIGMOD 2012 from the H-Store Project:

  • Skew-Aware Automatic Database Partitioning in Shared-Nothing, Parallel OLTP Systems
      Andrew Pavlo, Carlo Curino, Stanley Zdonik

    We present a novel approach to automatically partitioning databases for enterprise-class OLTP systems that significantly extends the state of the art by: (1) minimizing the number distributed transactions, while concurrently mitigating the effects of temporal skew in both the data distribution and accesses, (2) extending the design space to include replicated secondary indexes, (4) organically handling stored procedure routing, and (3) scaling of schema complexity, data size, and number of partitions. This effort builds on two key technical contributions: an analytical cost model that can be used to quickly estimate the relative coordination cost and skew for a given workload and a candidate database design, and an informed exploration of the huge solution space based on large neighborhood search. To evaluate our methods, we integrated our database design tool with a high-performance parallel, main memory DBMS and compared our methods against both popular heuristics and a state-of-the-art research prototype.

Accepted Papers

EDBT 2012 Accepted Paper

February 11th, 2012
Comments Off

The Brown Data Management Research Group has the following paper in EDBT 2012:

  • Optimizing Index Deployment Order for Evolving OLAP
      Hideaki Kimura, Carleton Coffrin, Alexander Rasin, Stanley Zdonik

    Many database applications need hundreds or thousands of indexes to speed up query execution, making performance tuning a difficult task for database administrators. The well-known problem of index selection is to automatically design an optimal set of indexes on behalf of DBAs. This paper brings a new perspective to the problem with a scalable and extensible solution. We study the problem of optimizing the order of index creation to achieve prompt query runtime improvements and also reduce the index deployment time. We found that traditional approaches, such as A* search and MIP, are not suitable for this problem. Instead, we demonstrate that Constraint Programming is an efficient and flexible solution for this problem and its future extension. Our experimental results show that our pruning techniques can reduce the size of the search space by many orders of magnitude, and we verify that our local search algorithm based on CP is a highly scalable and stable method for quickly finding the best known solutions.

Accepted Papers

ICDE 2012 Accepted Paper

December 12th, 2011
Comments Off

The Brown Data Management Research Group has the following paper in ICDE 2012:

  • Learning-based Query Performance Modeling and Prediction
       Mert Akdere, Ugur Cetintemel, Matteo Riondato, Eli Upfal, Stanley Zdonik (Project Longview)

    This paper studies the practicality and utility of sophisticated learning-based models for accurate query performance prediction (QPP). We propose and evaluate predictive modeling techniques that learn query execution behavior at different granularities, ranging from coarse-grained plan-level models to fine-grained operator-level models. We demonstrate that these two extremes offer a tradeoff between high accuracy for static workload queries and generality to unforeseen queries in dynamic workloads, respectively, and introduce a hybrid approach that combines their respective strengths by selectively composing them in the process of QPP. We discuss how we can use a training workload to (i) pre-build and materialize such models offline, so that they are readily available for future predictions, and (ii) build new models online as new predictions are needed. All prediction models are built using only static features (available prior to query execution) and the performance values obtained from the offline execution of the training workload. The results of an extensive experimental evaluation provide quantitative evidence that learning-based modeling for QPP is both feasible and effective for both static and dynamic workload scenarios.

Accepted Papers