Sérgio Lifschitz

Database Allocation Strategies for Parallel BLAST Evaluation on Clusters

In this work we investigate the parallel evaluation of BLAST, the most popular tool for comparing biological sequences. Our goal is to study database distribution issues that, besides workload balancing, could improve the performance of a set of BLAST processes running in a workstation cluster. We consider different partitioning strategies within actual BLAST executions against a few relevant molecular databases. We have implemented multiple databases and input sequence configurations and show that there are many important parameters, such as the fragment generation method and sequence similarities, that must be taken into account in order to make full use of such parallel environment.

A mechanism for replicated data consistency in mobile computing environments

Mobile computing allows for the development of new and sophisticated database applications. Such applications require the reading of current and consistent data. In order to improve data availability, increase performance and maximize throughput, data replication is used. However, due to inherent limitations in mobile and other loosely-coupled environments, the concurrency control and replica control mechanisms must be revisited. This paper proposes a new protocol that guarantees the consistency of replicated data in a mobile computing environment, while provide high data availability and ensure an eventual replica convergence towards a strongly consistent state.

Computing some distance functions between polygons

Abstract We present algorithms for computing some distance functions between two (possibly intersecting) polygons, in both the convex and nonconvex cases. The interest for such distance functions comes from applications in robot vision, pattern recognition and contour fitting. We present an optimal EREW-PRAM parallel algorithm for the case when the input polygons are convex, and an essentially quadratic sequential algorithm for the case of arbitrary polygons (possibly with holes).

Provenance in bioinformatics workflows.

In this work, we used the PROV-DM model to manage data provenance in workflows of genome projects. This provenance model allows the storage of details of one workflow execution, e.g., raw and produced data and computational tools, their versions and parameters. Using this model, biologists can access details of one particular execution of a workflow, compare results produced by different executions, and plan new experiments more efficiently. In addition to this, a provenance simulator was created, which facilitates the inclusion of provenance data of one genome project workflow execution. Finally, we discuss one case study, which aims to identify genes involved in specific metabolic pathways of Bacillus cereus, as well as to compare this isolate with other phylogenetic related bacteria from the Bacillus group. B. cereus is an extremophilic bacteria, collected in warm water in the Midwestern Region of Brazil, its DNA samples having been sequenced with an NGS machine.

A Genome Databases Framework

There are many Molecular Biology Databases, also known as Genome Databases, and there is a need for integrating all this data sources and related applications. This work proposes the use of an object-oriented framework for genome data access and manipulations. The framework approach is an interesting solution due to the flexibility, reusability and extensibility requirements of this application domain. We give a formal definition of our Genome Databases Framework using UML class diagrams, that explore the structural part of the architecture. A brief discussion on the Framework functionalities is also presented.

Implementation of an Agent Architecture for Automated Index Tuning

It has been important to extend database management systems to support new requirements of applications and administration. We focus in this paper on the automatic tuning feature, particularly manipulation of indexes. The architectural approach considered here is based on software agents. We briefly present the heuristics involved in the decision upon creation, dropping and the exact moment to proceed with either action. Then we discuss our agentbased database architecture, with its layers and coupling within an actual DBMS. Finally, we explain the way our hypothetical indexes are integrated in PostgreSQL and how they can be used in practice for automated index tuning.

Index Self-tunning with Agent-based Databases.

The use of software agents as Database Management System components lead to database systems that may be configured and extended to support new requirements. We focus here with the self-tuning feature, which demands a somewhat intelligent behavior that agents could add to traditional DBMS modules. We propose in this paper an agent-based database architecture to deal with automatic index creation. Implementation issues are also discussed, for a built-in agents and DBMS integration architecture.

A probabilistic view of Datalog parallelization

We explore an approach to developing Datalog parallelization strategies that aims at good expected rather than worst-case performance. To illustrate, we consider a very simple parallelization strategy that applies to all Datalog programs. We prove that this has very good expected performance under equal distribution of inputs. This is done using an extension of 0–1 laws adapted to this context. The analysis is confirmed by experimental results on randomly generated data.

Exploring Load Balancing in Parallel Processing of Recursive Queries

We study a dose-driven dynamic load balancing strategy to evaluate database recursive queries. This proposal aims at obtaining a good workload balance with full use of the available resources, with different kinds of skew considered. The strategy is intended for general recursive queries and preliminary computational results illustrate its efficiency when applied to the particular case of linear transitive closure.

Evaluating the Cassandra NoSQL Database Approach for Genomic Data Persistency.

Rapid advances in high-throughput sequencing techniques have created interesting computational challenges in bioinformatics. One of them refers to management of massive amounts of data generated by automatic sequencers. We need to deal with the persistency of genomic data, particularly storing and analyzing these large-scale processed data. To find an alternative to the frequently considered relational database model becomes a compelling task. Other data models may be more effective when dealing with a very large amount of nonconventional data, especially for writing and retrieving operations. In this paper, we discuss the Cassandra NoSQL database approach for storing genomic data. We perform an analysis of persistency and I/O operations with real data, using the Cassandra database system. We also compare the results obtained with a classical relational database system and another NoSQL database approach, MongoDB.