In which cases do system call invocations work faster: in PostgreSQL processes or in goroutines within the application?

System call challenges (syscalls) are executed within the kernel of the operating system and have roughly the same cost regardless of the context from which they are called. However, performance differences can arise due to application architecture.

• In PostgreSQL, each client request is handled by a separate process. A system call from a PostgreSQL process is a direct transition into the kernel.
• In Go, goroutines are lightweight threads scheduled within one or more system threads. A system call from a goroutine actually occurs from the system thread on which it runs.

The speed of a system call depends on how quickly the application switches between contexts and manages threads.

Conclusion:

• System calls from PostgreSQL processes usually do not have additional overhead related to the scheduler, as the process is a separate context.
• In Go, goroutines can switch between tasks faster, but the system call still executes in a system thread.

Thus, system calls from PostgreSQL processes and goroutines in Go are roughly comparable in speed. However, in Go, it is more efficient to manage a large number of parallel operations without creating many processes, which provides a scalability advantage but not a speed advantage for individual system calls.