You forgot to add a key difference between the two. In multiprocessing, we distribute the processes on different CPU cores so that the other cores are not idle and this is true multi-tasking. However, in multithreading, we synchronously execute the threads on the same CPU core and the threads wait for the CPU core to be idle to execute themselves. But as this happens fast, it looks like multi-tasking is happening. I believe this to be the most important difference.
Hey Vignesh, I disagree that in multithreading architecture threads always execute on single core. Multiple threads can execute on multiple cores giving you a true parallelism. Check this: stackoverflow.com/questions/11835046/multithreading-and-multicore-differences
Multithreading is light weight version of multiprocessing. You should use it as first preference. Multiple processes are littl difficult to manage but in python programming language due to GIL multithreading doesn't give you true parallelism, on this Occasion it makes sense to use multiprocessing
Appreciate this video, it was very digestible, and knowing that multiple threads exist in a single process, but they share the same address in memory that the process does, is exactly what I needed to know!
Step by step roadmap to learn data science in 6 months: th-cam.com/video/H4YcqULY1-Q/w-d-xo.html Full python course: th-cam.com/play/PLeo1K3hjS3usILfyvQlvUBokXkHPSve6S.html Machine learning tutorials with exercises: th-cam.com/video/gmvvaobm7eQ/w-d-xo.html
Thank you for a good video! I have one question, that I hope you can help me with. If we stay in you example with Power Point, what could a thread then be? What kind of tasks could it be? I hope you can help me with some examples.
@codebasics: but you said that processes can communicate with eachother via shared memories so if threads have a shared memory whats the difference if processes communicate via a shared memory too?
So Multiprocessing is effectively multitasking with applications (and other programs) And Multithreading is effectively multitasking within applications (and other programs) Makes a lot of sense. Edit: of course multiprocessing can also refer splitting tasks up between cores, but that can be used for either things.
How does hardware support for multiple threads in the same core improve performance when you use lots of threads in software? Let's say you want to perform 3D rendering and heavily use the CPU for several minutes or hours. I can see multiple cores helping a lot because each core could run each thread.
You really good at explanation. I only watched this video, and read comments. and its already seems obvious! Even accent isn't a problem when you know how to explain concept! Thanks! P.S. new feature of youtube with pointing to other videos in the end is also very good decision. It shows in some way that you are serious about channel and it also... bored to write:) Good job tho'.
3:30 -- well, the same counts for threads too, as long as you don't do something retarded on purpose, such as making two separate threads execute something on things stored within the same memory space.
I am here going through a home curriculum that is giving me headaches to understand github.andela.com/homestudy/ Please do more of this. I tend to understand concepts in such a way than reading.
When a process or compiler/interpreter doesn't free memory when suppose to or when the developer in the case that the memory is completely managed by the developer, the developer doesn't safely and or properly free the memory either upon uninitialization or when the memory is not in use at all.
Horrible explanation. After a 4 minute video, I still have no idea how is a process functionally different from threads. 3:30 was tasteless and disgusting as well. Disrespectful.
Kazathul. Every time that you start a program the operating system (Windows Linux, OSX) creates in the memory a restrict space of memory to store variables and the program by it self. If you open the notepad twice, they will run as two different processes for the MS Windows, for instance. In many (simple) computer programs you have a main loop or a sequence of instructions that you have to do. ----------------------------------------- var a, b, c; int main() { loop() { your code here } } --------------------------------------- To the above code if you start twice, the OS will create a copy of the code and the variables, as if you have two equals car in your garage. They are equal, but ... But, if you need to do a task, or more, in parallel to another, it is necessary to call delegate this to a specific code and this is possible in two ( ways), or you call a second program to do this or you call a thread. A second program for the specific task have their on space of memory their own code, but if comes with a cost. You have more problems to share information between this two tasks and you have a cost of memory. A way to do this more easily is to call a internal function of the program ans start a thread ----------------------------------------- int funcA() { } var a, b, c; int main() { // Goes to do something else t = thread('funcA'); t.start(); // Calls the function A in parallel // The last calls doesn't block the code and the loop starts loop() { your code here } } --------------------------------------- In a nonthread program if you call a funcA(), the main program it is blocked until the funcA() is finished, but in a thread program both is executed. Another point is that when a thread of the funcA() is executed, the variables a, b and c can be accessed. It doesn't happen in a separate process.
I absolutely love the clarity of this response. Cleared up the terms that I (probably we) were stuck on. AND the fucken savagery that accompanied the response. Absolutely deserved! lmfao 5 star response.
Do you want to learn technology from me? Check codebasics.io/ for my affordable video courses.
You forgot to add a key difference between the two. In multiprocessing, we distribute the processes on different CPU cores so that the other cores are not idle and this is true multi-tasking. However, in multithreading, we synchronously execute the threads on the same CPU core and the threads wait for the CPU core to be idle to execute themselves. But as this happens fast, it looks like multi-tasking is happening. I believe this to be the most important difference.
Hey Vignesh, I disagree that in multithreading architecture threads always execute on single core. Multiple threads can execute on multiple cores giving you a true parallelism. Check this: stackoverflow.com/questions/11835046/multithreading-and-multicore-differences
@@codebasics Python's Global Interpreter Lock enforces the rule Vighnesh mentioned above.
@@irina_divine you can bypass it pretty easily
@@codebasics o
For CPython (The most famous one), GIL prevents a thread from using multiple processors. I think @vighnesh153 is correct for CPython.
One of the best videos which clarifies distinction between process and threads
best memorizable visual examples so far. I pretty like your style to make the content understandable:D
Hi codebasics ,
When should I use multiprocessing?
When should I use multithreading ?
I hope u got my point.
Thanks,
Multithreading is light weight version of multiprocessing. You should use it as first preference. Multiple processes are littl difficult to manage but in python programming language due to GIL multithreading doesn't give you true parallelism, on this Occasion it makes sense to use multiprocessing
The best explanation so far!
Adorable concepts are clear in this video
Appreciate this video, it was very digestible, and knowing that multiple threads exist in a single process, but they share the same address in memory that the process does, is exactly what I needed to know!
Perfect Manu. You are an expert now 👍😊
Excellent explanation! Very easy to understand, saves time. Cheers to you
Can you please explain what is the meaning of the "processes are different programs written by different companies, so they have to run separately"?
Thank you Dhaval Sir.
That's the best explanation I've heard so far
that's actually the worst I have ever seen, the skinny and fat guys are stupid and confusing comparations.
Step by step roadmap to learn data science in 6 months: th-cam.com/video/H4YcqULY1-Q/w-d-xo.html
Full python course: th-cam.com/play/PLeo1K3hjS3usILfyvQlvUBokXkHPSve6S.html
Machine learning tutorials with exercises:
th-cam.com/video/gmvvaobm7eQ/w-d-xo.html
That was a perfect and great explonation! Thanks you very much sir!
This is seriously excellent content. Thank you.
Thank you so much for this video! It helped to clear a lot of my doubts on threads :)
I am happy this was helpful to you.
I'm so impressed by your explanations. Got a new sub.
Thank you for a good video! I have one question, that I hope you can help me with. If we stay in you example with Power Point, what could a thread then be? What kind of tasks could it be?
I hope you can help me with some examples.
@codebasics: but you said that processes can communicate with eachother via shared memories so if threads have a shared memory whats the difference if processes communicate via a shared memory too?
So Multiprocessing is effectively multitasking with applications (and other programs)
And Multithreading is effectively multitasking within applications (and other programs)
Makes a lot of sense.
Edit: of course multiprocessing can also refer splitting tasks up between cores, but that can be used for either things.
thank you for good explanation
What are some usecases/benefits of using Multithreading/MultiProcessing?
You did a great job in explaning the difference between these two... Keep it up
A great explanation thank you
Glad it was helpful!
How does hardware support for multiple threads in the same core improve performance when you use lots of threads in software? Let's say you want to perform 3D rendering and heavily use the CPU for several minutes or hours. I can see multiple cores helping a lot because each core could run each thread.
can you provide an example for memory leaks in threads ?
Is there a version without the background music? Thanks
Thank you! Great videos keep doing :)
best tutorials!! keep doing it
Thank you so much; that was a very good introduction into the topic; a very good video! :)
Such a great video... you cleared up so much! Thank You
Great video !!! Very informative
Great video. Whats the name of the song in the background?
Darude - Sandstorm
what is stack memory and instruction pointer ?
You really good at explanation. I only watched this video, and read comments. and its already seems obvious! Even accent isn't a problem when you know how to explain concept! Thanks!
P.S. new feature of youtube with pointing to other videos in the end is also very good decision. It shows in some way that you are serious about channel and it also...
bored to write:) Good job tho'.
Still don't know what multithreading is. Multiprocessing is simply cores doing multiple things at once
Thankyou, Sir !
Really a nice one.
very succinct and clear explanation ;)
thanks
do you have other channel name kudvenkat ?
3:30 -- well, the same counts for threads too, as long as you don't do something retarded on purpose, such as making two separate threads execute something on things stored within the same memory space.
yo is that the busy works beat theme song lol
thumbs up for the lightweight-heavyweight metaphor :D
please remove background music
That picture at 3:10 lol
I am here going through a home curriculum that is giving me headaches to understand github.andela.com/homestudy/
Please do more of this. I tend to understand concepts in such a way than reading.
Yeah. I just need someone to break the concepts down like the way you are doing. Will keep an eye out for them. Thanks.
V nice
Branded, thanks for the feedback
cool
Anyone in 2021
And what is a memory leak? :D
When a process or compiler/interpreter doesn't free memory when suppose to or when the developer in the case that the memory is completely managed by the developer, the developer doesn't safely and or properly free the memory either upon uninitialization or when the memory is not in use at all.
So what is her husband doing?
3:09💀
I dont speak english brou
Horrible explanation. After a 4 minute video, I still have no idea how is a process functionally different from threads.
3:30 was tasteless and disgusting as well. Disrespectful.
Kazathul lmaoo i feel you painnnn. Many videos and reading books, still can't understand wth that is. People never get straight to the point🤦🏿♀️
Kazathul. Every time that you start a program the operating system (Windows Linux, OSX) creates in the memory a restrict space of memory to store variables and the program by it self. If you open the notepad twice, they will run as two different processes for the MS Windows, for instance.
In many (simple) computer programs you have a main loop or a sequence of instructions that you have to do.
-----------------------------------------
var a, b, c;
int main()
{
loop()
{
your code here
}
}
---------------------------------------
To the above code if you start twice, the OS will create a copy of the code and the variables, as if you have two equals car in your garage. They are equal, but ...
But, if you need to do a task, or more, in parallel to another, it is necessary to call delegate this to a specific code and this is possible in two ( ways), or you call a second program to do this or you call a thread. A second program for the specific task have their on space of memory their own code, but if comes with a cost. You have more problems to share information between this two tasks and you have a cost of memory. A way to do this more easily is to call a internal function of the program ans start a thread
-----------------------------------------
int funcA()
{
}
var a, b, c;
int main()
{
// Goes to do something else
t = thread('funcA');
t.start(); // Calls the function A in parallel
// The last calls doesn't block the code and the loop starts
loop()
{
your code here
}
}
---------------------------------------
In a nonthread program if you call a funcA(), the main program it is blocked until the funcA() is finished, but in a thread program both is executed.
Another point is that when a thread of the funcA() is executed, the variables a, b and c can be accessed. It doesn't happen in a separate process.
I absolutely love the clarity of this response. Cleared up the terms that I (probably we) were stuck on. AND the fucken savagery that accompanied the response. Absolutely deserved! lmfao 5 star response.
you're hilarious kaza