Slide 1

Slide 1 text

@stevejgordon Turbocharged: Writing High-Performance C# and .NET Code @stevejgordon https://stevejgordon.co.uk Resources: http://bit.ly/highperfdotnet

Slide 2

Slide 2 text

@stevejgordon www.stevejgordon.co.uk • What is performance? • Measuring application and code performance • Span, ReadOnlySpan and Memory • ArrayPool • System.IO.Pipelines and ReadOnlySequence

Slide 3

Slide 3 text

@stevejgordon www.stevejgordon.co.uk Execution Time Throughput Memory Allocations

Slide 4

Slide 4 text

@stevejgordon Measure Optimise Measure Optimise OPTIMISATION CYCLE

Slide 5

Slide 5 text

@stevejgordon www.stevejgordon.co.uk • Visual Studio Diagnostic Tools (debugging) • Visual Studio Profiling / PerfView / dotTrace / dotMemory • ILSpy / JustDecompile / dotPeek / ILDASM • Production metrics and monitoring • Elastic APM Agent for .NET

Slide 6

Slide 6 text

@stevejgordon www.stevejgordon.co.uk • Library for .NET (micro)benchmarking • High precision measurements • Used extensively by .NET Runtime, CoreClr and ASP.NET Core teams https://benchmarkdotnet.org https://github.com/dotnet/BenchmarkDotNet

Slide 7

Slide 7 text

@stevejgordon www.stevejgordon.co.uk namespace BenchmarkExample { public class Program { public static void Main(string[] args) => _ = BenchmarkRunner.Run(); } [MemoryDiagnoser] public class NameParserBenchmarks { private const string FullName = "Steve J Gordon"; private static readonly NameParser Parser = new NameParser(); [Benchmark] public void GetLastName() { Parser.GetLastName(FullName); } } }

Slide 8

Slide 8 text

@stevejgordon www.stevejgordon.co.uk namespace BenchmarkExample { public class Program { public static void Main(string[] args) => _ = BenchmarkRunner.Run(); } [MemoryDiagnoser] public class NameParserBenchmarks { private const string FullName = "Steve J Gordon"; private static readonly NameParser Parser = new NameParser(); [Benchmark] public void GetLastName() { Parser.GetLastName(FullName); } } }

Slide 9

Slide 9 text

@stevejgordon www.stevejgordon.co.uk namespace BenchmarkExample { public class Program { public static void Main(string[] args) => _ = BenchmarkRunner.Run(); } [MemoryDiagnoser] public class NameParserBenchmarks { private const string FullName = "Steve J Gordon"; private static readonly NameParser Parser = new NameParser(); [Benchmark] public void GetLastName() { Parser.GetLastName(FullName); } } }

Slide 10

Slide 10 text

@stevejgordon www.stevejgordon.co.uk namespace BenchmarkExample { public class Program { public static void Main(string[] args) => _ = BenchmarkRunner.Run(); } [MemoryDiagnoser] public class NameParserBenchmarks { private const string FullName = "Steve J Gordon"; private static readonly NameParser Parser = new NameParser(); [Benchmark] public void GetLastName() { Parser.GetLastName(FullName); } } }

Slide 11

Slide 11 text

@stevejgordon www.stevejgordon.co.uk namespace BenchmarkExample { public class Program { public static void Main(string[] args) => _ = BenchmarkRunner.Run(); } [MemoryDiagnoser] public class NameParserBenchmarks { private const string FullName = "Steve J Gordon"; private static readonly NameParser Parser = new NameParser(); [Benchmark] public void GetLastName() { Parser.GetLastName(FullName); } } }

Slide 12

Slide 12 text

@stevejgordon www.stevejgordon.co.uk // * Summary * BenchmarkDotNet=v0.11.5, OS=Windows 10.0.18362 Intel Core i7-6700 CPU 3.40GHz (Skylake), 1 CPU, 8 logical and 4 physical cores .NET Core SDK=3.0.100 [Host] : .NET Core 3.0.0 (CoreCLR 4.700.19.46205, CoreFX 4.700.19.46214), 64bit RyuJIT DefaultJob : .NET Core 3.0.0 (CoreCLR 4.700.19.46205, CoreFX 4.700.19.46214), 64bit RyuJIT Method | Mean | Error | StdDev | Median | Gen 0 | Gen 1 | Gen 2 | Allocated | ------------ |-----------:|-----------:|-----------:|-----------:|-------:|-------:|-------:|----------:| GetLastName | 163.18 ns | 3.1903 ns | 4.2590 ns | 161.87 ns | 0.0379 | - | - | 160 B |

Slide 13

Slide 13 text

@stevejgordon www.stevejgordon.co.uk • System.Memory package. Built into .NET Core 2.1. • Provides a read/write 'view' onto a contiguous region of memory • Heap (Managed objects) – e.g. Arrays, Strings • Stack (via stackalloc) • Native/Unmanaged (P/Invoke) • Index / Iterate to modify the memory within the Span • Almost no overhead

Slide 14

Slide 14 text

@stevejgordon www.stevejgordon.co.uk Slicing a Span is a constant time/cost operation – O(1) Int[] myArray = new int[9] Span span1 = myArray.AsSpan() Span span2 = span1.Slice(start: 2, length: 5) Int[9] 0 1 2 3 4 5 6 7 8 0 1 2 3 4

Slide 15

Slide 15 text

OPTIMISING SOME CODE

Slide 16

Slide 16 text

Requirement: We need a method, that takes an array and returns ¼ of its elements, starting from the middle element.

Slide 17

Slide 17 text

var size = myArray.Length; myArray.Skip(size / 2). Take(size / 4).ToArray();

Slide 18

Slide 18 text

Requirement 2: Turbocharge it and prosper!!

Slide 19

Slide 19 text

@stevejgordon www.stevejgordon.co.uk [MemoryDiagnoser] public class ArrayBenchmarks { private int[] _myArray; [Params(100, 1000, 10000)] public int Size { get; set; } [GlobalSetup] public void Setup() { _myArray = new int[Size]; for (var i = 0; i < Size; i++) _myArray[i] = i; } // MORE CODE COMING RIGHT UP!!...

Slide 20

Slide 20 text

@stevejgordon www.stevejgordon.co.uk [MemoryDiagnoser] public class ArrayBenchmarks { private int[] _myArray; [Params(100, 1000, 10000)] public int Size { get; set; } [GlobalSetup] public void Setup() { _myArray = new int[Size]; for (var i = 0; i < Size; i++) _myArray[i] = i; } // MORE CODE COMING RIGHT UP!!...

Slide 21

Slide 21 text

@stevejgordon www.stevejgordon.co.uk [MemoryDiagnoser] public class ArrayBenchmarks { private int[] _myArray; [Params(100, 1000, 10000)] public int Size { get; set; } [GlobalSetup] public void Setup() { _myArray = new int[Size]; for (var i = 0; i < Size; i++) _myArray[i] = i; } // MORE CODE COMING RIGHT UP!!...

Slide 22

Slide 22 text

@stevejgordon www.stevejgordon.co.uk [MemoryDiagnoser] public class ArrayBenchmarks { // SETUP METHODS UP HERE! ... [Benchmark(Baseline = true)] public int[] Original() => _myArray.Skip(Size / 2).Take(Size / 4).ToArray(); ... }

Slide 23

Slide 23 text

@stevejgordon www.stevejgordon.co.uk | Method | Size | Mean | Ratio | Gen 0 | Gen 1 | Gen 2 | Allocated | |----------- |------ |---------------:|------:|-------:|-------:|------:|----------:| | Original | 100 | 154.9018 ns | 1.00 | 0.0534 | - | - | 224 B | | | | | | | | | | | Original | 1000 | 727.2669 ns | 1.00 | 0.2670 | - | - | 1120 B | | | | | | | | | | | Original | 10000 | 7,332.0136 ns | 1.00 | 2.4109 | - | - | 10120 B |

Slide 24

Slide 24 text

@stevejgordon www.stevejgordon.co.uk [MemoryDiagnoser] public class ArrayBenchmarks { ... [Benchmark] public int[] ArrayCopy() { var newArray = new int[Size / 4]; Array.Copy(_myArray, Size / 2, newArray, 0, Size / 4); return newArray; } ... }

Slide 25

Slide 25 text

@stevejgordon www.stevejgordon.co.uk | Method | Size | Mean | Ratio | Gen 0 | Gen 1 | Gen 2 | Allocated | |----------- |------ |---------------:|-------:|-------:|-------:|------:|----------:| | Original | 100 | 154.9018 ns | 1.000 | 0.0534 | - | - | 224 B | | ArrayCopy | 100 | 24.5267 ns | 0.159 | 0.0051 | - | - | 128 B | | | | | | | | | | | Original | 1000 | 727.2669 ns | 1.000 | 0.2670 | - | - | 1120 B | | ArrayCopy | 1000 | 104.7282 ns | 0.142 | 0.1627 | - | - | 1024 B | | | | | | | | | | | Original | 10000 | 7,332.0136 ns | 1.000 | 2.4109 | - | - | 10120 B | | ArrayCopy | 10000 | 801.1695 ns | 0.109 | 1.5917 | - | - | 10024 B |

Slide 26

Slide 26 text

@stevejgordon www.stevejgordon.co.uk [MemoryDiagnoser] public class ArrayBenchmarks { ... [Benchmark] public Span Span() => _myArray.AsSpan().Slice(Size / 2, Size / 4); ... }

Slide 27

Slide 27 text

@stevejgordon www.stevejgordon.co.uk | Method | Size | Mean | Ratio | Gen 0 | Gen 1 | Gen 2 | Allocated | |----------- |------ |---------------:|-------:|-------:|-------:|------:|----------:| | Original | 100 | 154.9018 ns | 1.000 | 0.0534 | - | - | 224 B | | ArrayCopy | 100 | 24.5267 ns | 0.159 | 0.0051 | - | - | 128 B | | Span | 100 | 0.9233 ns | 0.006 | - | - | - | - | | | | | | | | | | | Original | 1000 | 727.2669 ns | 1.000 | 0.2670 | - | - | 1120 B | | ArrayCopy | 1000 | 104.7282 ns | 0.142 | 0.1627 | - | - | 1024 B | | Span | 1000 | 0.9016 ns | 0.000 | - | - | - | - | | | | | | | | | | | Original | 10000 | 7,332.0136 ns | 1.000 | 2.4109 | - | - | 10120 B | | ArrayCopy | 10000 | 801.1695 ns | 0.109 | 1.5917 | - | - | 10024 B | | Span | 10000 | 0.9095 ns | 0.000 | - | - | - | - |

Slide 28

Slide 28 text

@stevejgordon www.stevejgordon.co.uk S ReadOnlySpan t e v e J G o r d o n ReadOnlySpan.Slice(start: 8) ReadOnlySpan span = "Steve J Gordon".AsSpan(); G o r d o n

Slide 29

Slide 29 text

@stevejgordon www.stevejgordon.co.uk • It's a stack only Value Type - ref struct • Requires C# >= 7.2 for ref struct feature • Cannot be a field in a class or standard (non ref) struct • Cannot be used as an argument or local variable inside async methods • Cannot be captured by lambda expressions

Slide 30

Slide 30 text

@stevejgordon www.stevejgordon.co.uk • Similar to Span but can live on the heap • A readonly struct but not a ref struct • Slightly slower to slice into Memory • Span property to get a Span over the same memory

Slide 31

Slide 31 text

@stevejgordon www.stevejgordon.co.uk // CS4012 Parameters or locals of type 'Span' cannot be declared // in async methods or lambda expressions. private async Task SomethingAsync(Span data) { ... // Would be nice to do something with the Span here await Task.Delay(1000); }

Slide 32

Slide 32 text

@stevejgordon www.stevejgordon.co.uk private async Task SomethingAsync(Memory data) { ... await Task.Delay(1000); }

Slide 33

Slide 33 text

@stevejgordon www.stevejgordon.co.uk private async Task SomethingAsync(Memory data) { Memory dataSliced = data.Slice(0, 100); await Task.Delay(1000); }

Slide 34

Slide 34 text

@stevejgordon www.stevejgordon.co.uk private async Task SomethingAsync(Memory data) { Memory dataSliced = data.Slice(0, 100); await Task.Delay(1000); } private void SomethingNotAsync(Span data) { // some code }

Slide 35

Slide 35 text

@stevejgordon www.stevejgordon.co.uk private async Task SomethingAsync(Memory data) { // CS4012 Parameters or locals of type 'Span' cannot be declared // in async methods or lambda expressions. var span = data.Span.Slice(1); SomethingNotAsync(span); await Task.Delay(1000); } private void SomethingNotAsync(Span data) { // some code }

Slide 36

Slide 36 text

@stevejgordon www.stevejgordon.co.uk private async Task SomethingAsync(Memory data) { SomethingNotAsync(data.Span.Slice(1)); await Task.Delay(1000); } private void SomethingNotAsync(Span data) { // some code }

Slide 37

Slide 37 text

@stevejgordon www.stevejgordon.co.uk Microservice which: 1. Reads an SQS message 2. Deserialise the JSON message 3. Stores a copy of the message to S3 using an object key derived from properties of the message. S3ObjectKeyGenerator

Slide 38

Slide 38 text

@stevejgordon www.stevejgordon.co.uk | Method | Mean [ns] | Ratio | Gen0 | Allocated [B] | Ratio | |------------- |----------:|---------:|-------:|--------------:|------:| | Original | 790.1 ns | | 0.1154 | 728 B | | | SpanBased | 386.5 ns | -51% | 0.0305 | 192 B | -74% | | StringCreate | 310.9 ns | -60% | 0.0305 | 192 B | -74% | >2x Faster ~3.8x Less Allocations 18 million messages: Reduction of 9.65GB of allocations daily Removes approx. 1528 Gen 0 collections

Slide 39

Slide 39 text

@stevejgordon www.stevejgordon.co.uk • Pool of arrays for re-use • Found in System.Buffers • ArrayPool.Shared.Rent(int length) • You are likely to get an array larger than your minimum size • ArrayPool.Shared.Return(T[] array, bool clearArray = false) • Warning! By default, returned arrays are not cleared

Slide 40

Slide 40 text

@stevejgordon www.stevejgordon.co.uk public class Processor { public void DoSomeWorkVeryOften() { var buffer = new byte[1000]; // allocates DoSomethingWithBuffer(buffer); } private void DoSomethingWithBuffer(byte[] buffer) { // use the array } }

Slide 41

Slide 41 text

@stevejgordon www.stevejgordon.co.uk public class Processor { public void DoSomeWorkVeryOften() { var buffer = new byte[1000]; // allocates DoSomethingWithBuffer(buffer); } private void DoSomethingWithBuffer(byte[] buffer) { // use the array } }

Slide 42

Slide 42 text

@stevejgordon www.stevejgordon.co.uk public class Processor { public void DoSomeWorkVeryOften() { var arrayPool = ArrayPool.Shared; var buffer = arrayPool.Rent(1000); DoSomethingWithBuffer(buffer); } private void DoSomethingWithBuffer(byte[] buffer) { // use the array } }

Slide 43

Slide 43 text

@stevejgordon www.stevejgordon.co.uk public class Processor { public void DoSomeWorkVeryOften() { var arrayPool = ArrayPool.Shared; var buffer = arrayPool.Rent(1000); try { DoSomethingWithBuffer(buffer); } finally { arrayPool.Return(buffer); } } private void DoSomethingWithBuffer(byte[] buffer) { // use the array } }

Slide 44

Slide 44 text

@stevejgordon www.stevejgordon.co.uk • Originally created by ASP.NET team to improve Kestrel rps • Improves I/O performance scenarios (~2x vs. streams) • Removes common hard to write, boilerplate code • Unlike streams, pipelines manages buffers for you from the ArrayPool • Two ends to a pipe, a PipeWriter and a PipeReader

Slide 45

Slide 45 text

@stevejgordon www.stevejgordon.co.uk PipeWriter : IBufferWriter Pipe PipeReader Memory m = pw.GetMemory(); … pw.Advance(1000) await pw.FlushAsync() ReadResult r = await reader.ReadAsync(); ReadOnlySequence b = r.Buffer;

Slide 46

Slide 46 text

@stevejgordon www.stevejgordon.co.uk Memory Memory Memory ReadOnlySequence

Slide 47

Slide 47 text

@stevejgordon www.stevejgordon.co.uk Microservice which: 1. Retrieves S3 object (TSV file) from AWS 2. Decompresses file 3. Parses TSV to get 3 of 25 columns for each row 4. Indexes data to Elasticsearch CloudFrontParser

Slide 48

Slide 48 text

@stevejgordon www.stevejgordon.co.uk | Method | Mean |Ratio | Gen 0 | Gen 1 | Gen 2 | Allocated |Ratio | |---------- |----------:|-----:|---------:|---------:|---------:|----------:|-----:| | Original | 47.47 ms | - | 14090.91 | 3272.73 | 1454.55 | 100.68 MB | - | | Optimised | 10.86 ms | -77% | 546.87 | 531.25 | 15.63 | 3.35 MB | -97% | Processing 1 file of 10,000 rows ~30x Less Heap Memory Allocations NOTE: ~2.85MB are the string allocations for the parsed data. Overhead = 0.5MB

Slide 49

Slide 49 text

@stevejgordon www.stevejgordon.co.uk •Identify a quick win •Use a scientific approach to demonstrate gains •Put gains into a monetary value •Cost to benefit ratio

Slide 50

Slide 50 text

@stevejgordon www.stevejgordon.co.uk This work is a small part of a much bigger potential gain For a single microservice handling 18 million messages per day Reduction of at least 50% of allocations. At least 1 less VM needed per year saving $1,700 Potential to at least double per instance throughput

Slide 51

Slide 51 text

@stevejgordon www.stevejgordon.co.uk A single (micro)service could save $1,700. These gains can scale with additional (micro)services. $17,000?? $170,000???

Slide 52

Slide 52 text

@stevejgordon www.stevejgordon.co.uk • Measure, don't assume! • Be scientific; make small changes each time and measure again • Focus on hot paths • Don't copy memory, slice it! Span is less complex than it may first seem. • Use ArrayPools where appropriate to reduce array allocations • Consider Pipelines for I/O scenarios

Slide 53

Slide 53 text

@stevejgordon www.stevejgordon.co.uk Pro .NET Memory Management By Konrad Kokosa

Slide 54

Slide 54 text

@stevejgordon www.stevejgordon.co.uk @stevejgordon | stevejgordon.co.uk http://bit.ly/highperfdotnet