Global operation across all workers
res = gop(FUN,x)
res = gop(FUN,x,targetlab)
| Function to operate across workers. |
| Argument to function |
| Variable to hold reduction result. |
| Lab to which reduction results are returned. This value
is returned by that worker’s |
res = gop(FUN,x) is the reduction
via the function FUN of the quantities x from
each worker. The result is duplicated on all workers.
FUN can be a handle to any function, including
user-written functions and user-defined anonymous functions. It should
accept two arguments of the same type, and return one result of that
same type, so it can be used iteratively in the form:
FUN(FUN(x1,x2),FUN(x3,x4))
The function FUN should be associative, that
is,
FUN(FUN(x1,x2),x3) = FUN(x1,FUN(x2,x3))
res = gop(FUN,x,targetlab) performs the
reduction, and places the result into res only
on the worker indicated by targetlab. res is
set to [ ] on all other workers.
This example shows how to calculate the sum and maximum values
for x among all workers.
p = parpool('local',4); x = Composite(); x{1} = 3; x{2} = 1; x{3} = 4; x{4} = 2; spmd xsum = gop(@plus,x); xmax = gop(@max,x); end xsum{1}
10
xmax{1}4
This example shows how to horizontally concatenate the column
vectors of x from all workers into a matrix. It
uses the same 4-worker parallel pool opened by the previous example.
x{1} = [3;30];
x{2} = [1;10];
x{3} = [4;40];
x{4} = [2;20];
spmd
res = gop(@horzcat,x);
end
res{1} 3 1 4 2
30 10 40 20
This example shows how to use an anonymous function with gop to
join character vectors with spaces between them. In this case, the
character vectors are created from each worker’s labindex value.
afun = @(a,b)[a,' ',b] spmd res = gop(afun,num2str(labindex)); end res{1}
1 2 3 4