Поиск  
Always will be ready notify the world about expectations as easy as possible: job change page
Nov 20, 2022

Super Fast Query in Entity Framework

Автор:
Juldhais Hengkyawan
Источник:
Просмотров:
2994

Sometimes we have to perform several queries to the database, put the results into memory, and then process them as needed.

For example, we have a Supplier table and a Customer table. Both tables have a substantial amount of data. We want to combine existing Supplier and Customer data with specific criteria.

Generally, we will perform two queries sequentially: query to the Supplier table first, then query to the Customer table. The second query will be executed after completing the first query.

If each query takes 5 seconds to process, and the query merge process in memory takes 2 seconds, then the total time required to complete the entire process is about 12 seconds (5+5+2).

Now the question arises: “Can we perform the first and second query simultaneously (parallel), then combine the results after the two queries are finished?”

The answer is: YES, WE CAN!

The total time required to complete the entire process will be significantly reduced by running the first and second queries simultaneously.

How?

First, create a Console Application project with the .NET 6.0 framework. Name the project AsyncEntity.

Add the EntityFrameworkCore.SqlServer package from NuGet Packages to the project.

Then create two entities, Supplier and Customer:

namespace AsyncEntity;

public class Supplier
{
    public int Id { get; set; }
    public string Name { get; set; }
}

 

namespace AsyncEntity;

public class Customer
{
    public int Id { get; set; }
    public string Name { get; set; }
}

Next, we create a data context class. Create a new file named DataContext.cs:

using Microsoft.EntityFrameworkCore;

namespace AsyncEntity;

public class DataContext : DbContext
{
    public DataContext(DbContextOptions options) : base(options)
    {
    }

    public DbSet<Supplier> Supplier { get; set; }
    public DbSet<Customer> Customer { get; set; }
}

We need to populate the Supplier and Customer tables with dummy data. Create the DataInitializer class to initialize the data:

namespace AsyncEntity;

public class DataInitializer
{
    private readonly DataContext _db;

    public DataInitializer(DataContext db)
    {
        _db = db;
    }

    public async Task Run()
    {
        await _db.Database.EnsureDeletedAsync();
        await _db.Database.EnsureCreatedAsync();

        var total = 1000000;
        for (int i = 0; i < total; i++)
        {
            _db.Supplier.Add(new Supplier
            {
                Name = $"P{(i + total)}"
            });

            _db.Customer.Add(new Customer
            {
                Name = $"C{(i + total)}"
            });
        }

        await _db.SaveChangesAsync();
    }
}

When the Run() method in the DataInitializer class is executed, the Program will reset the database (EnsureDeleted() + EnsureCreated()) and fill the Supplier and Customer tables with 1 million data.

It’s time to run the data initialization process in Program.cs:

using System.Diagnostics;
using AsyncEntity;
using Microsoft.EntityFrameworkCore;

var stopwatch = new Stopwatch();

var connectionString = "Data Source=.\\SQL2019;Initial Catalog=AsyncEntity;Integrated Security=True";
var optionsBuilder = new DbContextOptionsBuilder().UseSqlServer(connectionString);

// initialize data
stopwatch.Start();
var dbInitializer = new DataContext(optionsBuilder.Options);
var dataInitializer = new DataInitializer(dbInitializer);
await dataInitializer.Run();

stopwatch.Stop();
Console.WriteLine($"Data initialization: {stopwatch.Elapsed}");

The data initialization process took about 1 minute 23 seconds on my computer.

Data initialization: 00:01:23.9350880

Sequential

Next, we’ll create a sequential process first:

// sequential
stopwatch.Restart();
var dbSequential = new DataContext(optionsBuilder.Options);

var sequentialSuppliers = await dbSequential.Supplier
    .Where(x => x.Name.Contains("7"))
    .OrderByDescending(x => x.Name)
    .ToListAsync();

var sequentialCustomers = await dbSequential.Customer
    .Where(x => x.Name.Contains("7"))
    .OrderByDescending(x => x.Name)
    .ToListAsync();

var sequentialTotal = sequentialSuppliers.Count + sequentialCustomers.Count;
stopwatch.Stop();
Console.WriteLine($"Sequential (Total: {sequentialTotal}): {stopwatch.Elapsed}");

In the code snippet above, the query is still executed sequentially, even though we have done asynchronous programming using the ToListAsync() method.

Sequential (Total: 937118): 00:00:04.5931360

The sequential process above takes about 4.5 seconds.

Parallel

Next, we will make the first query and the second query run in parallel:

// parallel
stopwatch.Restart();

// create two DataContext instance
var dbParallel1 = new DataContext(optionsBuilder.Options);
var dbParallel2 = new DataContext(optionsBuilder.Options);

// task for the first query (no need to await here)
var parallelSupplierTask = dbParallel1.Supplier    
    .Where(x => x.Name.Contains("7"))
    .OrderByDescending(x => x.Name)
    .ToListAsync();

// task for the second query (no need to await here)
var parallelCustomerTask = dbParallel2.Customer
    .Where(x => x.Name.Contains("7"))
    .OrderByDescending(x => x.Name)
    .ToListAsync();

// run both tasks in parallel (await here)
await Task.WhenAll(parallelSupplierTask, parallelCustomerTask);

// get the results from both queries
var parallelSuppliers = parallelSupplierTask.Result;
var parallelCustomers = parallelCustomerTask.Result;

var parallelTotal = parallelSuppliers.Count + parallelCustomers.Count;
stopwatch.Stop();
Console.WriteLine($"Parallel (Total: {parallelTotal}): {stopwatch.Elapsed}");

The code snippet above will run the first and second queries in parallel and wait until both queries are finished.

Parallel (Total: 937118): 00:00:02.6591467

The parallel process above only takes about 2.6 seconds! Almost half of the sequential process.

By using Task.WhenAll(), we can run multiple tasks at the same time. This will undoubtedly increase the performance and speed of the Program we make.

However, we must be careful because we have to use a different data context for each query that will be executed in parallel. Multiple data contexts require more memory allocation and more connection pool.

The source code of this article can be found here: https://github.com/juldhais/AsyncEntity

Похожее
Jun 27, 2023
Author: Anton Selin
Introduction Performance optimization is a key concern in software development, regardless of the programming language or platform you’re using. It’s all about making your software run faster or with less memory consumption, leading to better user experience and more efficient...
Aug 17, 2023
Author: Geoff Graham
What does your performance “stack” look like? There are all kinds of tools available for measuring page speed, but what data and assumptions do they use to measure performance? And speaking of measuring performance, there’s quite a difference between that...
Jun 10
Author: Dayanand Thombare
LINQ (Language Integrated Query) has revolutionized the way we interact with data in C#. It offers a consistent, readable, and concise way to manipulate collections, databases, XML, and more. However, the beauty and ease of LINQ can sometimes mask performance...
Apr 8
Author: João Simões
Performance comparison between LinkedList and ToArray Some weeks ago I created an article comparing the performance of ToList versus ToArray when creating short lived collections that won’t be mutated, usually used to prevent multiple enumerations when iterating over a temporary...
Написать сообщение
Тип
Почта
Имя
*Сообщение