A random collection of performance tests I've done using Benchmark.NET. To be very clear, please, do not take these tests and their results and make incorrect conclusions. Meaning, just because my tests show that technique A is better than B, it doesn't mean that is a global truism. The results.md file contains results for each test, but again, take that information as it is, and don't necessarily use that data to make sweeping, production-level changes in your code.
Following are brief descriptions of each set of tests.
This is a test to see what happens when a null check is added for a parameter to a method.
This is a test to compare looking up a method by either a well-known name or by the existence of an attribute.
This looks at using a Dictionary<,> over a switch or if.
This is a test to see if there's any difference in assigining property values via "traditional" setting and tuples.
This is a test to see how much cost is involved in using dynamic to call a method.
This is a test to see the cost in using dynamic for property usage.
This is a test to see the differences in calculating the n-th number in the Fibonacci sequence.
This is a test to compare using anonymous and tuple types.
This is a test to see what happens when the capacity is set when a List<> is created.
This is a test to compare ways to negate a number. (If one wonders why I did this, I faintly remember running into the string parsing technique at a client I was working at, and I was trying to show that maybe there was a faster way to do it, along with handling negative numbers correctly.)
This tests different ways to order conditional statements.
This is a test to see what checked does to performance.
This tested "interesting" Guid parsing code I found at a client. The service and controller layers are mimicked in the test.
This looks to see if calling Contains() before Replace() helps.
This tests different ways to get the sum of all the elements in a list.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent