How do I use a parallel.foreach?

Question

0.00/5 (No votes)

See more:

I am trying to use a parallel.foreach. However, I noticed that when I set a count to see the iteration, I noticed that the amount of times that it iterates through is less than the actual amount of iterations.

Here is my code:

Parallel.ForEach(dt.AsEnumerable,
       Sub(f)
           countI += 1
       End Sub)

count of dt = 154563
count of parallel.foreach = 154510

can someone please help me with this. I need to iterate through all of them through parallel foreach.

What I have tried:

Tried to change it to a parallel.for but this made the count 1/3 instead which is worse.

Posted 28-Dec-19 9:22am

Member 11856456

Updated 28-Dec-19 19:40pm

Add a Solution

2 solutions

Solution 2

As Griff pointed out parallel.foreach can get very problematic.
An alternative which works well (in my experience) is using Tasks, which is also recommended by Microsoft, see example here: TaskFactory.StartNew Method (System.Threading.Tasks) | Microsoft Docs[^]

Posted 28-Dec-19 19:40pm

RickZeeland

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

OriginalGriff · Accepted Answer · 2019-12-28T10:38:00

The problem is that when you try to parallelise "basic code" you have to be carefull about what you are doing.

C#

x += 1

is "syntactic sugar", it's actually this:

C#

x = x + 1

Which in "machine terms" is a short sequence of discrete operations:

1) Load "x" into a register.
2) Add one to the register.
3) Load the register into "x".

When you parallelise that sequence, you get multiple tasks running it as close to simultaneously as possible, and because you don't have 154,563 separate cores in your processor that means that they don't all execute the same steps at the same time. So if you have one core that has just loaded "x" into a register, and a second core that is just about to load a new value into it, then the first task is going to overwrite the second tasks revised value with it's new one - but since they both loaded the same value the new value ends up incremented once only!

This gets complicated, and there are ways to avoid it, but they effectively "force" all the parallel task to execute sequentially because the whole of the method you wrote has to be treated as a single discrete operation and that loses any advantage of parallel operation! (In fact, it'll actually slow the whole thing down considerably, as each thread needs it's own memory space, stack, thread controller, and will require a significant amount of task switching work by the operating system.