Thank you! The comment stated at time 10:42 is probably one of the most important parts! Essentially, one thread (within a given grid) will create/allocate the shared array and all other threads within the same grid will ignore that line of code and assume that it has been handled by the first thread.
im gonna give you a subscribe. keep up the good work .sometimes i am really suprised how rare it is to find an actual good introduction to computer science related topics.
Extremely clear explanation of different GPU mem, cleaning the spurious dust on GPU. Really useful info for a better GPU memory management. THX from UY!
Hello Sir! I have a doubt ... wouldn't the transfer of data from cpu to gpu (global memory) be the bottle-neck in the entire architecture? You teach great! Thank You !!!
Thanks for the content of this video very well explained. One issue arises about the __shared__ memory access; I guess there is a mistake in the proposed program: while the index i=threadIdx.x + blockDim.x * blockIdx.x is used to access the "IN" array stored in device memory the shared_array instead should not be indexed by i but by "int j = threadIdx.x". If not then N (size of the array) must be large as the size of the block or less (as max as 1024 threads). If N is larger than the size of the block then errors about accessing violation could rise up. Do you agree?
![Seungmin "그렇게, 천천히, 우리(As we are)" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/kAzmhLHePqU/mqdefault.jpg)
WOW !!! THE BEST EXPLANATION EVER !!! SO CLEAR !!!
Excellent explanation of GPU memory hierarchy. Good job!
This is the best CUDA tutorial on the internet!!!
Thank you! The comment stated at time 10:42 is probably one of the most important parts! Essentially, one thread (within a given grid) will create/allocate the shared array and all other threads within the same grid will ignore that line of code and assume that it has been handled by the first thread.
This was astoundingly good. Thank you so much for making these lectures and for releasing them to the public
im gonna give you a subscribe. keep up the good work .sometimes i am really suprised how rare it is to find an actual good introduction to computer science related topics.
extremly good explanation of complex topic in an easy way.. You are awsome
Extremely clear explanation of different GPU mem, cleaning the spurious dust on GPU. Really useful info for a better GPU memory management. THX from UY!
Great stuff! Everything I needed to form a decent mental model of a GPU
The best CUDA tutorial !!!
Fantastic explanation! Looking forward to future content if you decide to produce more. Much appreciated so far!
This is much better than I read the cuda c book...
this was so helpful, feel like I need to do more than just subscribe 😀
Great content indeed!!! I'm willing to pay for such content 🔥
This series is great! How come such a video has only 4K views after over 2 years? Is everybody on a Python hype-train these days? ;)
Maybe not much people use cuda.
These are great.
And they're pre-covid. What other good lectures were released online pre-pandemic?!
Brilliant!
Thank you
Hello Sir! I have a doubt ... wouldn't the transfer of data from cpu to gpu (global memory) be the bottle-neck in the entire architecture? You teach great! Thank You !!!
Thank you sir for the great lectures, how can we access the rest of the course, is there more than those 6 parts?
look up the books he's showing at the end of the video
Thanks for the content of this video very well explained.
One issue arises about the __shared__ memory access; I guess there is a mistake in the proposed program: while the index i=threadIdx.x + blockDim.x * blockIdx.x is used to access the "IN" array stored in device memory the shared_array instead should not be indexed by i but by "int j = threadIdx.x". If not then N (size of the array) must be large as the size of the block or less (as max as 1024 threads). If N is larger than the size of the block then errors about accessing violation could rise up. Do you agree?
Concise and Precise
thx!
what's the difference between Cuda core and thread?
very clear,learn
CUDA 101
cant understand your english ascent..😢
This is the best CUDA tutorial on the internet!!!