performance - Estimate parallel efficiency using unicore processor -


We know that parallel efficiency of a program running on a multicore system can be calculated as speed / N Where N is the number is definitely So to use this formula first we need to execute the code on a multicore system and need to know the speed

I would like to know that I do not have a multicore system, so what is possible Can the speed of code given on a multicore system be estimated by executing it on a Unicorn processor?

I have access to the functionality counter (instructions per cycle, number of caches, number of instructions, etc.) and I only have the binaries of code.

[Note: I guessed

P>, especially about the bit.

To determine how much you can speed up your program, you need to know how some parts of the program can be benefited by parallel and which part should be executed sequentially To If you know that, and if you know that when you take serial and parallel parts (personally) on a processor, you can guess how fast the program will be on many processors.

From your description, it seems that you do not know which parts can use parallel processing and which parts to be executed sequentially. So it will not be possible to estimate the running parallel time.

Comments

Popular posts from this blog

c - Mpirun hangs when mpi send and recieve is put in a loop -

python - Apply coupon to a customer's subscription based on non-stripe related actions on the site -

java - Unable to get JDBC connection in Spring application to MySQL -