[julia-users] Need help writing parallel code with @sync and @everywhere

[Daniel Carrera]

Hello,

I have been having a lot of trouble figuring out how to write parallel code 
in Julia. Consider this toy example of a serial program:

    N = 5
    for i = 1:N
        for j = (i + p):N
            println(i * j)
        end
    end

Now, suppose that "i * j" is an expensive operation, so I want to compute 
those values in parallel and print them later. Here is my (failed) attempt 
at doing that:

    N = 5
    tmp = SharedArray(Int, (N))
    
    for i = 1:N
        # ----------------------- #
        # Compute tmp in parallel #
        # ----------------------- #
        @sync begin
            @everywhere begin
                p = myid()
                np = nprocs()
                for j = (i + p):np:N
                    tmp[j] = i * j
                end
            end
        end
        
        # --------------------- #
        # Consume tmp in serial #
        # --------------------- #
        for j = (i + 1):N
            println(tmp[j])
        end
    end


So, my idea is to have a shared array with N elements where I store some of 
the "i * j" calculations in parallel. Once everyone is finished, I consume 
the results and repeat the loop. However, I get an error saying that "i" is 
not visible inside the @everywhere block:

julia> nprocs()
4

julia> foo_parallel()
exception on 1: ERROR: i not defined
 in anonymous at /home/sigrid/Daniel/Science/VENUS/venus.jl:303
 in eval at /build/buildd/julia-0.3.8-docfix/base/sysimg.jl:7
 in anonymous at multi.jl:1310
 in run_work_thunk at multi.jl:621
 in run_work_thunk at multi.jl:630
 in anonymous at task.jl:6
 ... many lines ...


So, apparently the workers in the @everywhere block cannot see outside 
variables. What can I do? ... One thing I do not want to do is turn "tmp" 
into an N x N matrix. That would make "tmp" very large for large N, and in 
my real problem there are other outside variables that I want to use.

Help?

Cheers,
Daniel.