How CUDA Programming Works | GTC 2022

[@sami9323]

this is one of the clearest and most lucid presentations i have seen, on any topic

[@zoriiginalx7544]

The illustration of how RAM works on a physical level really was illuminating. Really drove home why linear accesses are important and why cache lines are the way they are. Fantastic talk.

[@dennisrkb]

Great presentation on GPU architecture, performance tradeoffs and considerations.

[@ypwangreg]

I was always puzzled and fascinated about how those grid/block/threads work in parallel in the GPU and this video explains it in one and all. very impressive and helpful!

[@SrikarDurgi]

Dan is definitely the MAN. Great talk!

[@holeo196]

Another great presentation by Stephen Jones, fascinating

[@KingDestrukto]

Fantastic presentation, wow!

[@hadiorabi692]

Man this is amazing

[@KalkiCharcha-hd5un]

@21:17 "Its exactly the right amount of data to hit the peak bandwidth of my mem system , Even if my program reads data from all over the place , each read is exactly ONE page of my memory " I didnt understand this statement 21:17 "Even if my program reads data from all over the place" Does it mean even if the data is read from non consecutive memory ??

[@perli216]

yes

[@perli216]

You got the benefits of reading contiguous memory for free basically, even when doing random reads

[@KalkiCharcha-hd5un]

@@perli216 Ok cool so basically only mem is contigues we get advantage like if i = tid + bid*bsize , and not like i = 2*(tid + bid*bsize)

[@perli216]

@@KalkiCharcha-hd5un I don't understand your question

[@KalkiCharcha-hd5un]

@@perli216 "Even if my program reads data from all over the place" , I think I got it , Initially I thought "... all over the place" as in any random memory / non consecutive . all over the place as in diff threads from same page , because single thread will bring in the data from same page anyway.

[@miramar-103]

superb!

[@chamidou2023]

Great presentation!

[@purandharb]

thanks for detailed explanation. Really enjoyed it.

[@kimoohuang]

Great presentation！It is mentioned that 4 warps x 256 bytes per warp = 1024 bytes, and that equals to the Memory page size 1024 bytes. It only happens when the 4 warps running adjacent threads。Are the 4 warps always running adjacent threads?

[@perli216]

@@kimoohuang Not necessarily. Depends on the warp scheduler

[@mugglepower]

oh man I hope my mum fixed me with a better brain processing unit so I could understand this

[@LetoTheSecond0]

Looks like the link in the description is broken/truncated?

[@perli216]

@@LetoTheSecond0 yes, yourube did this. It's just the original source for the video

[@openroomxyz]

Interesting!