unnused - mergen-sergio/DBest GitHub Wiki

Key Takeaways

  • Manual Query Plan Creation: DBest requires the user to create the query plan manually, using a visual interface or through other tools that DBest provides.
  • No Query Optimization: Unlike traditional DBMSs, DBest does not automatically analyze or optimize query plans.
  • Query Execution and Cost Reporting: DBest focuses on executing the user-defined query plan and provides useful cost indicators to help users evaluate the performance of their queries.

Exporting Result Sets

In DBest, the result sets produced by the query operators can be exported in two formats:

  1. CSV Format
  2. Proprietary B+ Tree Format

These export options provide flexibility depending on how you want to use or store the resulting data after query execution.

The Proprietary B+ Tree export option allows the result set to be saved in DBest's optimized B+ Tree format. This format is ideal for use cases where you need to perform efficient range queries or lookups on the exported data.

  • Indexing: The B+ Tree format indexes a set of columns, which enables efficient searches and range queries. When exporting to the B+ Tree format, users can define the key columns that will be indexed for quick access.
  • Use Case: This is particularly useful when preparing data once, so it can later be reused for fast, repeated queries that require sorting, equality checks, or range searches.

Table of Contents

  1. Overview
  2. Features
  3. Installation
  4. Usage
  5. Contributing
  6. License
  7. Acknowledgments

Example Workflow

  1. Step 1: Open a CSV for a large dataset and define the datatypes of each column.
  2. Step 2: Create a query plan over the CSV file to select the columns of interest
  3. Step 3: Export the result set as a B+ tree, indexing specific columns (e.g., Age, Salary) for efficient querying. Export multiple times if different indexes are needed.
  4. Step 4: Prepare a B+ Tree from a large dataset, indexing specific columns (e.g., Age, Salary) for efficient querying.
  5. Step 5: Open the B+ Tree and used it in multiple queries to perform range queries, sorting, or lookups without rebuilding the index each time.

Dependências externas

Nome Descrição
For Your Information Database (FYI Database) Utilizado como o banco de dados principal
ANTLR4 Utilizado para validar as entradas possíveis da DSL desenvolvida para o software
JGraphX Utilizado para construir os nós visuais e suas conexões
Data Faker Utilizado para gerar dados e popular as tabelas criadas na ferramenta

Contribuições

Caso você queira contribuir com o projeto, abaixo estão os contatos das pessoas que participam dele.

Nome Contato
Sérgio [email protected]
Rhuan [email protected] [email protected]
Luiz [email protected]
Marcos [email protected] [email protected]