Building Microservices with Go

Profiling

When we wish to take a look at the speed of our program, the best technique we can employ is profiling. Profiling automatically samples your running application while it is executing; and then we can compute that data, such as the running time of a particular function, into a statistical summary called a profile.

Go supports three different types of profiling:

• CPU: Identifies the tasks which require the most CPU time
• Heap: Identifies the statements responsible for allocating the most memory
• Blocking: Identifies the operations responsible for blocking Go routines for the longest time

If we would like to enable profiling on our application, we can do one of two things:

• Add import "net/http/pprof" to your startup file
• Manually start ...