C#并发请求异常计时报告

Question

我正在开发一个向 API 发送多个请求的控制台应用程序，我正在使用异步、任务和等待。我正在使用秒表来显示每个 request/task 花费的时间，我注意到它开始时非常低（150 毫秒），并且每个下一个任务都会增加大约 100 毫秒。我认为这些任务是运行ning 并发的，因为程序在 8 秒内完成了 83 requests/tasks，当我用 Chrome 测量 get 请求时，它显示大约 200 毫秒。

你知道为什么时间随着任务的进行而增加吗？测量或我的代码逻辑有问题吗？这不是应该更快吗？据我所知，WhenAll 应该运行任务并发，总完成时间是列表中的最大任务时间。

public async Task<List<CatalogEvent>> GetEventsAsync(int id)
    {
        sw.Restart();
        var request = GetRequest(msCatalogEndpoint);
        request.AddParameter("id", id, ParameterType.UrlSegment);
        List<CatalogEvent> events = new List<CatalogEvent>();
        var response = await client.ExecuteTaskAsync(request).ConfigureAwait(false);
        var catalog = JsonConvert.DeserializeObject<CatalogEndpoint>(response.Content);
        if (!(catalog.catalogEvents is null))
        {
            foreach (var ev in catalog.catalogEvents)
            {
                CatalogEvent catalogEvent = ev.Value;
                catalogEvent.eventName = ev.Key.ToString();
                catalogEvent.titleId = id;
                DateTime dateTime = DateTime.UtcNow;
                catalogEvent.date = dateTime.ToString();
                events.Add(catalogEvent);
            }
        }
        Console.WriteLine($"Task for Id: {id} took {sw.ElapsedMilliseconds} ms and was managed by Thread: {Thread.CurrentThread.ManagedThreadId}");
        return events;
    }

我正在使用 RestSharp 包来发出请求。主要方法是这样的：

        static void Main(string[] args)
    {
        //this list has 83 ids which I am getting from a database
        List<int> ids = GetIds();

        async Task ProcessEvents()
        {
            IEnumerable<Task<List<CatalogEvent>>> techBriefEvents = ids.Select(id => GetEventsAsync(id));
            await Task.WhenAll(techBriefEvents);
        }

        Task.WhenAll(ProcessEvents());
        Console.ReadKey();
    }

这是输出：

Task for TitleId: 142 took 164 ms and was managed by Thread: 8
Task for TitleId: 16 took 349 ms and was managed by Thread: 5
Task for TitleId: 10 took 634 ms and was managed by Thread: 6
Task for TitleId: 215 took 650 ms and was managed by Thread: 5
Task for TitleId: 114 took 826 ms and was managed by Thread: 6
Task for TitleId: 214 took 843 ms and was managed by Thread: 5
Task for TitleId: 56 took 983 ms and was managed by Thread: 6
Task for TitleId: 212 took 1001 ms and was managed by Thread: 5
Task for TitleId: 168 took 1141 ms and was managed by Thread: 6
Task for TitleId: 21 took 1168 ms and was managed by Thread: 5
Task for TitleId: 26 took 1309 ms and was managed by Thread: 6
Task for TitleId: 30 took 1334 ms and was managed by Thread: 5
Task for TitleId: 213 took 1462 ms and was managed by Thread: 6
Task for TitleId: 24 took 1510 ms and was managed by Thread: 5
Task for TitleId: 29 took 1619 ms and was managed by Thread: 6
Task for TitleId: 23 took 1669 ms and was managed by Thread: 5
Task for TitleId: 31 took 1779 ms and was managed by Thread: 6
Task for TitleId: 14 took 1906 ms and was managed by Thread: 5
Task for TitleId: 18 took 1943 ms and was managed by Thread: 6
Task for TitleId: 20 took 2064 ms and was managed by Thread: 6
Task for TitleId: 19 took 2110 ms and was managed by Thread: 6
Task for TitleId: 175 took 2222 ms and was managed by Thread: 8
Task for TitleId: 15 took 2275 ms and was managed by Thread: 6
Task for TitleId: 102 took 2400 ms and was managed by Thread: 8
Task for TitleId: 33 took 2464 ms and was managed by Thread: 8
Task for TitleId: 135 took 2563 ms and was managed by Thread: 5
Task for TitleId: 5 took 2632 ms and was managed by Thread: 8
Task for TitleId: 137 took 2750 ms and was managed by Thread: 5
Task for TitleId: 12 took 2796 ms and was managed by Thread: 8
Task for TitleId: 41 took 2911 ms and was managed by Thread: 5
Task for TitleId: 136 took 2998 ms and was managed by Thread: 8
Task for TitleId: 43 took 3084 ms and was managed by Thread: 5
Task for TitleId: 139 took 3159 ms and was managed by Thread: 8
Task for TitleId: 51 took 3240 ms and was managed by Thread: 5
Task for TitleId: 42 took 3322 ms and was managed by Thread: 5
Task for TitleId: 39 took 3393 ms and was managed by Thread: 5
Task for TitleId: 44 took 3502 ms and was managed by Thread: 8
Task for TitleId: 122 took 3583 ms and was managed by Thread: 5
Task for TitleId: 36 took 3697 ms and was managed by Thread: 8
Task for TitleId: 95 took 3744 ms and was managed by Thread: 5
Task for TitleId: 67 took 3871 ms and was managed by Thread: 8
Task for TitleId: 229 took 3896 ms and was managed by Thread: 5
Task for TitleId: 226 took 4034 ms and was managed by Thread: 8
Task for TitleId: 108 took 4078 ms and was managed by Thread: 5
Task for TitleId: 123 took 4213 ms and was managed by Thread: 8
Task for TitleId: 143 took 4285 ms and was managed by Thread: 5
Task for TitleId: 236 took 4364 ms and was managed by Thread: 8
Task for TitleId: 228 took 4466 ms and was managed by Thread: 5
Task for TitleId: 232 took 4540 ms and was managed by Thread: 6
Task for TitleId: 230 took 4641 ms and was managed by Thread: 5
Task for TitleId: 149 took 4715 ms and was managed by Thread: 6
Task for TitleId: 176 took 4793 ms and was managed by Thread: 5
Task for TitleId: 208 took 4902 ms and was managed by Thread: 6
Task for TitleId: 155 took 4946 ms and was managed by Thread: 5
Task for TitleId: 61 took 5057 ms and was managed by Thread: 6
Task for TitleId: 190 took 5097 ms and was managed by Thread: 5
Task for TitleId: 93 took 5262 ms and was managed by Thread: 5
Task for TitleId: 194 took 5280 ms and was managed by Thread: 5
Task for TitleId: 156 took 5419 ms and was managed by Thread: 6
Task for TitleId: 101 took 5440 ms and was managed by Thread: 5
Task for TitleId: 193 took 5572 ms and was managed by Thread: 6
Task for TitleId: 167 took 5598 ms and was managed by Thread: 5
Task for TitleId: 197 took 5730 ms and was managed by Thread: 6
Task for TitleId: 111 took 5755 ms and was managed by Thread: 5
Task for TitleId: 216 took 5882 ms and was managed by Thread: 6
Task for TitleId: 60 took 5930 ms and was managed by Thread: 5
Task for TitleId: 9 took 6059 ms and was managed by Thread: 5
Task for TitleId: 152 took 6085 ms and was managed by Thread: 5
Task for TitleId: 169 took 6218 ms and was managed by Thread: 6
Task for TitleId: 154 took 6264 ms and was managed by Thread: 5
Task for TitleId: 7 took 6403 ms and was managed by Thread: 6
Task for TitleId: 141 took 6506 ms and was managed by Thread: 5
Task for TitleId: 58 took 6560 ms and was managed by Thread: 6
Task for TitleId: 172 took 6670 ms and was managed by Thread: 5
Task for TitleId: 11 took 6730 ms and was managed by Thread: 6
Task for TitleId: 17 took 6846 ms and was managed by Thread: 5
Task for TitleId: 55 took 6912 ms and was managed by Thread: 6
Task for TitleId: 166 took 7020 ms and was managed by Thread: 5
Task for TitleId: 140 took 7069 ms and was managed by Thread: 6
Task for TitleId: 110 took 7177 ms and was managed by Thread: 5
Task for TitleId: 90 took 7222 ms and was managed by Thread: 6
Task for TitleId: 160 took 7352 ms and was managed by Thread: 5
Task for TitleId: 97 took 7400 ms and was managed by Thread: 6
Task for TitleId: 200 took 7503 ms and was managed by Thread: 5
Task for TitleId: 153 took 7556 ms and was managed by Thread: 6
Task for TitleId: 207 took 7654 ms and was managed by Thread: 5
Task for TitleId: 161 took 7721 ms and was managed by Thread: 6
Task for TitleId: 231 took 7810 ms and was managed by Thread: 5
Task for TitleId: 202 took 7873 ms and was managed by Thread: 6
Task for TitleId: 220 took 8068 ms and was managed by Thread: 6

Answer 1

一个明显的误解是"and was managed by Thread:"。除非ExecuteTaskAsync执行的很烂，there is no thread.

如果向同一台主机发出请求，您可能会运行进入 service point manager 限制。

C#并发请求异常计时报告

C# concurrent requests strange timing reported

c#

http

stopwatch

request

async-await