SQL Engine

The SQL Query Execution Engine is a critical component of any Database Management System (DBMS). It plays a key role in processing SQL queries and transforming them into meaningful results. Every time a user runs a query to fetch data from a database, various complex processes occur simultaneously to retrieve the desired output.

In this article, we will explain the phases of SQL query execution, the roles of different components within the query execution process, and how different DBMS tools such as MySQL, SQL Server, and others provide an interface to interact with the SQL Engine.

What Happens When a Query Is Executed?

When we submit an SQL query, the database engine processes it in multiple stages. The query first undergoes parsing, followed by optimization, execution planning, and finally, execution. During this process, the query is converted into relational algebra expressions and then executed in a sequence that ensures optimal performance and accurate results.          

Example Query

Select * from account
Where balance>1000;

The aim of this query is to find out details of all accounts where the balance is greater than 1000. In relational algebra, this query can be expressed as:

Relational Algebra Expression 1:

πbalance( δbalance>1000(account) ) 

Relational Algebra Expression 2:

δbalance>1000( πbalance(account) ) 

Both of the query expressions give the same result, but they demonstrate the transformation from SQL to relational algebra. Once the query is parsed then generally two things happen:

1. Multiple equivalent action plans are generated.
2. A query order tree is created for each action plan, and the query order tree is evaluated.

Phases of SQL Query Evaluation

When we execute an SQL query, it goes through several important phases that ensure the efficient retrieval of data. These phases include Parsing, Optimization, Execution Plan Creation, and the final phase of Execution by the Query Execution Engine. Let’s explore these phases in detail.

1. Parser/Translator

The first phase of SQL query evaluation is parsing, which is handled by the Parser/Translator. The primary job of the parser is to:

• Verify the Syntax: It checks whether the SQL query follows correct syntax.
• Check Semantics: It ensures that the query is semantically correct, i.e., all table names, column names, and operators used in the query are valid.
• Convert to Relational Algebra: Since SQL is a high-level language, it needs to be translated into a lower-level language that the system understands. The parser translates the SQL query into relational algebra expressions, which form the foundation for further query optimization and execution.

2. Optimizer

The Optimizer is an important component that improves the efficiency of query execution. Once the query is parsed into relational algebra, the optimizer selects the best execution plan from multiple possible alternatives. The optimizer uses various data statistics and indexing information to decide which relational algebra expression will minimize resource usage, such as CPU time and disk I/O.

How Optimizers Work:

• Cost-based optimization: The optimizer estimates the cost of each potential query plan based on factors like available indexes, join methods, and query selectivity.
• Selection of Low-Cost Plans: The optimizer selects the most efficient plan that will return results with minimal resource consumption.

3. Execution Plan

The Execution Plan is where the Database Engine decides the exact order of operations for executing the query. The order of operations can significantly affect performance. For example, consider the following query:

Select * From emp_table
Where experience>5;

The typical execution order is:

1. FROM – The database engine first identifies the source table.
2. WHERE – It applies the filtering conditions (e.g., experience > 5).
3. SELECT – Finally, the database engine retrieves the required columns.

This step-by-step order of execution is crucial in ensuring that the query is performed efficiently.

4. Query Execution Engine

The execution engine will perform operations by fetching data from the database. The Execution Engine reads the Execution Plan and interprets each step, executing the query accordingly. It interacts directly with the storage layer to retrieve and manipulate the data, and then returns the final output.

Functions of the Query Execution Engine:

• Dispatcher: It serves as a dispatcher for all operations in the execution plan, performing actions like fetching data from storage, applying filters, and joining tables.
• Interaction with Storage Engine: The Execution Engine directly communicates with the Storage Engine to retrieve data from tables and indexes. It also updates data when necessary.
• Execution of Complex Queries: Modern DBMS tools such as Apache Spark, Presto, and Apache Drill include advanced query execution engines that support distributed computing and complex query processing.

SQL Query Plan Example

SELECT * FROM employee
WHERE department = 'HR' AND salary > 50000;

Explanation:

1. Check the table: Identify if there are any indexes on the department or salary columns.
2. Access the data: Retrieve rows where department = 'HR' and salary > 50000.
3. Return the results: The results are then returned to the user.

How SQL Database Engines Handle Data

SQL database engines use various data structures to manage and access data efficiently. The most common data structures used are B-trees (balanced trees) and hash tables, which allow quick access to data by indexing key values.

• Storage Engine: Manages how data is stored and retrieved from the disk.
• Relational Engine: Handles the query optimizer, execution, and relational algebra translation.
• Execution Engine: Executes the final operations and returns the results.

SQL Engines in Different DBMS Tools

Most relational database management systems (RDBMS), including MySQL, SQL Server, PostgreSQL, and Oracle, provide an interface and APIs to interact with the SQL engine. The SQL engine works in the background to interpret and execute SQL queries efficiently. Here’s a brief overview of how various DBMS tools use SQL engines:

1. MySQL SQL Engine

MySQL uses the InnoDB Storage Engine by default, which efficiently handles transaction management and row-level locking. The MySQL Query Execution Engine optimizes queries by choosing the best access path based on available indexes and data distribution.

In MySQL, we can use the following command to check available storage engines:

SHOW ENGINES;

This will display all the available engines (e.g., InnoDB, MyISAM) and their features.

2. SQL Server Query Engine

SQL Server features a Query Optimizer that works with its SQL Execution Engine to determine the best execution plan for a query. It uses a cost-based optimizer to select the optimal strategy for executing queries based on system resources and data availability.

3. PostgreSQL Query Engine

PostgreSQL is known for its advanced query planning and optimization techniques. It uses a planner to determine the best query execution path, relying on a cost-based system to generate the most efficient query plan.

4. Oracle SQL Engine

Oracle uses the Oracle Optimizer and Execution Engine to optimize and execute SQL queries. It is highly customizable and uses a variety of techniques, including index usage, partition pruning, and parallel execution for complex queries.

Conclusion

The SQL query execution engine is fundamental to the performance and efficiency of database management systems. By parsing, optimizing, and executing SQL queries, the engine ensures that data retrieval is fast and resource-efficient. Understanding how SQL engines work can help developers write better queries, optimize performance, and troubleshoot issues effectively. Whether we are using MySQL, SQL Server, PostgreSQL, or any other RDBMS, all SQL engines follow similar principles but have their own unique optimizations.