If all you want is minimal overlapping, you can just give each of your threads an "index" (from 0, 1, 2, ..., n - 1 where n is your thread count) and make each thread aware of how many threads it is working with. Then, each thread with index T does work unit T, T + n, T + 2n, T + 3n, ... there you have it, zero overlapping. For two threads, the first one will do 0, 2, 4, 6, ... and the second one will do 1, 3, 5, 7, ... all within the same "function".

Note you probably don't want to use System.out.println when working with threads, since they'll print their stuff in nondeterministic order depending on which thread got there first. You want to keep a memory buffer which'll store your thread's results (one slot for each work unit you're doing) and each thread writes there. Then, at the end, you can process it however you want.

Also, for multithreading to be of any use, you need your function to contain significant work that can be parallelized. If you're just trying to get two threads to somehow execute an arbitrary function "together" (your diagram is rather vague), that's not the way to go, especially considering each thread has its own stack and CPU context.

I find OpenMP to be rather refreshing when it comes to getting familiar with the concepts involved in scalable multithreading. Don't know if Java has that though.

If all you want is minimal overlapping, you can just give each of your threads an "index" (from 0, 1, 2, ..., n - 1 where n is your thread count) and make each thread aware of how many threads it is working with. Then, each thread with index T does work unit T, T + n, T + 2n, T + 3n, ... there you have it, zero overlapping. For two threads, the first one will do 0, 2, 4, 6, ... and the second one will do 1, 3, 5, 7, ... all within the same "function".

Note you probably don't want to use System.out.println when working with threads, since they'll print their stuff in nondeterministic order depending on which thread got there first. You want to keep a memory buffer which'll store your thread's results (one slot for each work unit you're doing) and each thread writes there. Then, at the end, you can process it however you want.

Also, for multithreading to be of any use, you need your function to contain significant work that can be parallelized. If you're just trying to get two threads to somehow execute an arbitrary function in lockstep (your diagram is rather vague), that's not the way to go, especially considering each thread has its own stack and CPU context.

I find OpenMP to be rather refreshing when it comes to getting familiar with the concepts involved in scalable multithreading. Don't know if Java has that though.

If all you want is minimal overlapping, you can just give each of your threads an "index" (from 0, 1, 2, ..., n - 1 where n is your thread count) and make each thread aware of how many threads it is working with. Then, each thread with index T does work unit T, T + n, T + 2n, T + 3n, ... there you have it, zero overlapping. For two threads, the first one will do 0, 2, 4, 6, ... and the second one will do 1, 3, 5, 7, ... all within the same "function".

Note you probably don't want to use System.out.println when working with threads, since they'll print their stuff in nondeterministic order depending on which thread got there first. You want to keep a memory buffer which'll store your thread's results (one slot for each work unit you're doing) and each thread writes there. Then, at the end, you can process it however you want.

I find OpenMP to be rather refreshing when it comes to getting familiar with the concepts involved in scalable multithreading. Don't know if Java has that though.

If all you want is minimal overlapping, you can just give each of your threads an "index" (from 0, 1, 2, ..., n - 1 where n is your thread count) and make each thread aware of how many threads it is working with. Then, each thread with index T does work unit T, T + n, T + 2n, T + 3n, ... there you have it, zero overlapping. For two threads, the first one will do 0, 2, 4, 6, ... and the second one will do 1, 3, 5, 7, ... all within the same "function".

Note you probably don't want to use System.out.println when working with threads, since they'll print their stuff in nondeterministic order depending on which thread got there first. You want to keep a memory buffer which'll store your thread's results (one slot for each work unit you're doing) and each thread writes there. Then, at the end, you can process it however you want.