Limiting Concurrent Operations With SemaphoreSlim Using C#
If you’re looking to limit the number of concurrent operations but maintain as high throughput as possible SemaphoreSlim can help! For instance, this could help maintain a consistent flow of HTTP requests to an external API during a bulk processing operation – respecting the limits of the external API to not potentially overwhelm it with too many concurrent requests.
Example
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
using System.Threading;
using System.Threading.Tasks;
namespace MyApp
{
public class MyService
{
private const int MaximumConcurrentOperations = 10;
private static readonly SemaphoreSlim RateLimit = new SemaphoreSlim(
initialCount: MaximumConcurrentOperations,
maxCount: MaximumConcurrentOperations);
public async Task Process()
{
await RateLimit.WaitAsync();
try
{
// Rate limited logic here
}
finally
{
RateLimit.Release();
}
}
}
}
maxCount
If maxCount is not specified int.MaxValue is used. From my understanding maxCount in this context is sort of a safety mechanism, making sure you aren’t releasing more than waiting.
Consider the following example:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
using System;
using System.Collections.Generic;
using System.Threading;
using System.Threading.Tasks;
namespace MyApp
{
class Program
{
static readonly SemaphoreSlim SemaphoreSlim =
new SemaphoreSlim(initialCount: 1);
static void Main(string[] args)
{
var tasks = new List<Task>();
for (var i = 0; i < 10; i++)
{
tasks.Add(Task.Run(() =>
{
SemaphoreSlim.Wait();
Console.WriteLine($"Start task, CurrentCount: {SemaphoreSlim.CurrentCount}");
Thread.Sleep(100);
Console.WriteLine($"End task, CurrentCount: {SemaphoreSlim.CurrentCount}");
SemaphoreSlim.Release();
SemaphoreSlim.Release(); // Extra call to Release();
}));
}
Task.WaitAll(tasks.ToArray());
}
}
}
With example output:
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Start task, CurrentCount: 0
End task, CurrentCount: 0
Start task, CurrentCount: 1
Start task, CurrentCount: 0
End task, CurrentCount: 0
End task, CurrentCount: 0
Start task, CurrentCount: 1
Start task, CurrentCount: 0
Start task, CurrentCount: 0
Start task, CurrentCount: 1
End task, CurrentCount: 0
End task, CurrentCount: 0
End task, CurrentCount: 0
End task, CurrentCount: 0
Start task, CurrentCount: 1
Start task, CurrentCount: 0
Start task, CurrentCount: 1
End task, CurrentCount: 5
End task, CurrentCount: 5
End task, CurrentCount: 5
However, if the SemaphoreSlim is changed to new SemaphoreSlim(initialCount: 1, maxCount: 1) then System.Threading.SemaphoreFullException is thrown instead.
Summary
Rate limiting using SemaphoreSlim can help avoid overwhelming external services with too many concurrent requests while maintaining high throughput.