Ensure Thatparfor-Loop Iterations are Independent

If you get an error when you convertfor-loops toparfor-loops, ensure that yourparfor-loop iterations are independent.parfor-loop iterations haveno guaranteed order, while the iteration order infor-loops issequential. Alsoparfor-loop iterations are performed on different MATLAB^®workers in the parallel pool, so that there is no sharing of information between iterations. Therefore oneparfor-loop iteration must not depend on the result of a previous iteration. The only exception to this rule is to accumulate values in a loop usingReduction Variables.

The following example produces equivalent results, using afor-loop on the left and aparfor-loop on the right. Try the example in your MATLAB Command Window:

clearAfori = 1:8 A(i) = i;endA

A = 1 2 3 4 5 6 7 8

clearAparfori = 1:8 A(i) = i;endA

A = 1 2 3 4 5 6 7 8

Each element ofAis equal to its index. Theparfor循环工作因为se each element is determined by the indexed loop variable only and does not depend on other variables.for-loops with independent tasks are ideal candidates forparfor-loops.

In the example, the array elements are available in the client workspace after theparfor-loop, exactly as with afor-loop.

Now use a nonindexed variable inside the loop, or a variable whose indexing does not depend on the loop variablei. Try these examples, and note the values ofdandiafterward:

clearAd = 0; i = 0;fori = 1:4 d = i*2; A(i) = d;endA d i

A = 2 4 6 8 d = 8 i = 4

clearAd = 0; i = 0;parfori = 1:4 d = i*2; A(i) = d;endA d i

A = 2 4 6 8 d = 0 i = 0

Although the elements ofAare the same in both examples, the value ofdis not. In thefor-loop, the iterations are executed sequentially, so afterwarddhas the value it held in the last iteration of the loop. In theparfor-loop, however, the iterations execute in parallel, so it is impossible to assignda defined value at the end of the loop. This situation also applies to the loop variablei. Therefore,parfor循环行为定义,以便它不affect the valuesdandioutside the loop. Their values remain the same before and after the loop. If the variables in yourparfor-loop are not independent, then you might get different answers from those in thefor-loop. In summary, aparfor-loop requires that each iteration be independent of the other iterations. All code that follows theparforstatement should not depend on the loop iteration sequence.

Code Analyzer can help diagnose whether the loop iterations are dependent. The code in the example shows iterations defined in terms of the previous iteration:

parfork = 2:10 x(k) = x(k-1) + k;end

In other cases, however, Code Analyzer is unable to mark dependencies.

For help with other commonparforproblems, seeNested parfor and for-Loops and Other parfor Requirements.

See Also

parfor

Related Examples

• Decide When to Use parfor
• Convert for-Loops Into parfor-Loops
• Nested parfor and for-Loops and Other parfor Requirements
• Troubleshoot Variables in parfor-Loops
• Reduction Variables

More About

• Evaluate Functions in the Background Using parfeval