专门针对密集运算而优化的Channel
多路写入 单路读取
特点:
内存连续,cpu命中率高
自旋锁,解锁时间更敏捷,可以自行定制锁
使用Span<T>, 执行效率更好, 还可以使用LinqFaster,NetFabric.Hyperlinq等高效的linq
gc less
// * Summary *
BenchmarkDotNet=v0.13.1, OS=macOS Big Sur 11.6 (20G165) [Darwin 20.6.0] Intel Core i9-9880H CPU 2.30GHz, 1 CPU, 16 logical and 8 physical cores [Host] : Mono 6.12.0.125 (2020-02/8c552e98bd6 Mon), X64 Mono : Mono 6.12.0.125 (2020-02/8c552e98bd6 Mon), X64
Job=Mono Runtime=Mono
| Method | inSleep | outSleep | Mean | Error | StdDev | Rank | Gen 0 | Allocated |
|---|---|---|---|---|---|---|---|---|
| _2_FixedChannel_SpinLock | 0 | 0 | 194.5 us | 1.44 us | 1.35 us | 1 | - | - |
| _2_FixedChannel_Lock | 0 | 0 | 22,001.3 us | 84.85 us | 79.36 us | 4 | - | - |
| _2_SystemChannel_Unbounded | 0 | 0 | 577.0 us | 2.18 us | 1.82 us | 2 | 2.9297 | - |
| _2_SystemChannel_Bounded | 0 | 0 | 1,175.7 us | 14.68 us | 13.73 us | 3 | 5.8594 | - |
| _2_FixedChannel_SpinLock | 0 | 1 | 288,271.9 us | 308.64 us | 288.70 us | 5 | - | - |
| _2_FixedChannel_Lock | 0 | 1 | 290,593.0 us | 2,089.37 us | 1,954.40 us | 5 | - | - |
| _2_SystemChannel_Unbounded | 0 | 1 | 304,186.5 us | 820.11 us | 767.13 us | 6 | - | - |
| _2_SystemChannel_Bounded | 0 | 1 | 304,744.6 us | 1,537.57 us | 1,438.24 us | 6 | - | - |
| _2_FixedChannel_SpinLock | 1 | 0 | 288,223.9 us | 525.16 us | 491.24 us | 5 | - | - |
| _2_FixedChannel_Lock | 1 | 0 | 301,892.3 us | 775.19 us | 725.11 us | 6 | - | - |
| _2_SystemChannel_Unbounded | 1 | 0 | 303,623.2 us | 518.50 us | 485.01 us | 6 | - | - |
| _2_SystemChannel_Bounded | 1 | 0 | 303,165.8 us | 1,937.80 us | 1,812.62 us | 6 | - | - |
| _2_FixedChannel_SpinLock | 1 | 1 | 290,501.4 us | 232.50 us | 217.48 us | 5 | - | - |
| _2_FixedChannel_Lock | 1 | 1 | 303,851.1 us | 994.79 us | 930.53 us | 6 | - | - |
| _2_SystemChannel_Unbounded | 1 | 1 | 302,817.6 us | 574.29 us | 537.20 us | 6 | - | - |
| _2_SystemChannel_Bounded | 1 | 1 | 302,535.3 us | 779.03 us | 728.71 us | 6 | - | - |