kelvinrfr/netcore-benchmarking

netcore-benchmarking

Repository to hold some benchmark tests using the https://github.com/dotnet/BenchmarkDotNet library.

Content

1. Perform benchmark of some code
2. Perform some unit tests over performance
3. Add docker support and run benchmarks over benchmark classes

   dotnet tool install -g BenchmarkDotNet.Tool
   dotnet benchmark MyAssemblyWithBenchmarks.dll --filter *

1. Perform benchmark of some code

Testing the following methods, where N is the size of Data list.

void LinqWhereAndCount() => Data.Where(x => x % 2 == 0).Count();
void LinqCount() => Data.Count(x => x % 2 == 0);

This code is on Benchmark_Code.Program. The results are the following:

N = 10

Method	Mean	Error	StdDev
LinqWhereAndCount	108.6 ns	1.77 ns	1.66 ns
LinqCount	184.6 ns	3.64 ns	4.86 ns

N = 100

Method	Mean	Error	StdDev
LinqWhereAndCount	504.2 ns	9.35 ns	13.70 ns
LinqCount	1,867.5 ns	35.35 ns	33.07 ns

N = 10,000

Method	Mean	Error	StdDev
LinqWhereAndCount	99.48 us	1.928 us	2.295 us
LinqCount	197.62 us	3.847 us	5.393 us

N = 50,000

Method	Mean	Error	StdDev
LinqWhereAndCount	497.7 us	8.80 us	8.23 us
LinqCount	982.3 us	13.70 us	12.81 us

N = 100,000

Method	Mean	Error	StdDev
LinqWhereAndCount	974.1 us	19.27 us	18.02 us
LinqCount	1,937.9 us	38.37 us	53.79 us

Conclusion

This shows that is better use a .Where(...).Count() instead of .Count(...) as the second option has a increase on the processing time when the list size increases.

Legends

• Mean : Arithmetic mean of all measurements
• Error : Half of 99.9% confidence interval
• StdDev : Standard deviation of all measurements
• 1 ns : 1 Nanosecond (0.000000001 sec)
• 1 us : 1 Microsecond (0.000001 sec)