Prática de Shadowing: Nvidia CUDA in 100 Seconds - Aprenda a falar inglês com vídeo

Criando lição...
1
CUDA, a parallel computing platform that allows you to use your GPU for more than just playing video games.
2
Compute Unified Device Architecture was developed by NVIDIA in 2007 based on the prior work of Ian Buck and John Nichols.
3
Since then, CUDA has revolutionized the world by allowing humans to compute large blocks of data in parallel, which has unlocked the true potential of the deep neural networks behind artificial intelligence.
4
The Graphics Processing Unit, or GPU, is historically used for what the name implies, to compute graphics.
5
When you play a game in 1080p at 60fps, you've got over 2 million pixels on the screen that may need to be recalculated after every frame,
6
which requires hardware that can do a lot of matrix multiplication and vector transformations in parallel.
7
And I mean a lot.
8
Modern GPUs are measured in teraflops, or how many trillions of floating point operations can it handle per second?
9
Unlike modern CPUs like the Intel i9, which has 24 cores, a modern GPU like the RTX 4090 has over 16,000 cores.
10
A CPU is designed to be versatile, while a GPU is designed to go really fast in parallel.
11
CUDA allows developers to tap into the GPU's power, and data scientists all around the world are using at this very moment, trying to train the most powerful machine learning models.
12
It works like this.
13
You write a function, called a CUDA kernel, that runs on the GPU.
14
You then copy some data from your main RAM over to the GPU's memory, then the CPU will tell the GPU to execute that function or kernel in parallel.
15
The code is executed in a block, which itself organizes threads into a multi-dimensional grid.
16
Then the final result from the GPU is copied back to the main memory.
17
Piece of cake, let's go ahead and build a CUDA application right now.
18
First you'll need an NVIDIA GPU, then install the CUDA toolkit.
19
CUDA includes device drivers, runtime, compilers, and dev tools, but the actual code is most often written in C++, as I'm doing here in Visual Studio.
20
First, we use the global specifier to define a function or CUDA kernel that runs on the actual GPU.
21
This function adds two vectors or arrays together.
22
It takes pointer arguments A and B, which are the two vectors to be added together, and pointer C for the result.
23
C equals A plus B, but because hypothetically we're doing billions of operations in parallel, we need to calculate the global index of the thread in the block that we're working on.
24
From there, we can use managed, which tells CUDA this data can be accessed from both the host CPU and the device GPU, without the need to manually copy data between them.
25
And now we can write a main function for the CPU that runs the CUDA kernel.
26
We use a for loop to initialize our arrays with data, then from there, we pass this data to the add function to run it on the GPU.
27
But you might be wondering what these weird triple brackets are.
28
They allow us to configure the CUDA kernel launch to control how many blocks
29
and how many threads per block are used to run this code in parallel.
30
And that's crucial for optimizing multi-dimensional data structures like tensors used in deep learning.
31
From there, CUDA Device Synchronize will pause the execution of this code and wait for it to complete on the GPU.
32
When it finishes and copies the data back to the host machine, we can then use the result and print it to the standard output.
33
Now, let's execute this code with a CUDA compiler by clicking the play button.
34
Congratulations, you just ran 256 threads in parallel on your GPU.
35
But if you want to go beyond, NVIDIA's GTC conference is coming up in a few weeks.
36
It's free to attend virtually, featuring talks about building massive parallel systems with CUDA.
37
Thanks for watching, and I will see you in the next one.
📺 Mesmo canal
10 very promising Open Source Projects you haven’t heard ofB2
1

10 very promising Open Source Projects you haven’t heard of

American Accent, en-USUniversity Students+2
Array Map in 100 SecondsC1
1

Array Map in 100 Seconds

American Accent, en-USUniversity Students+2
7 Programming myths that waste your timeB2
1

7 Programming myths that waste your time

American Accent, en-USUniversity Students+2
The plan to break apart Google... RIP ChromeC1
5

The plan to break apart Google... RIP Chrome

American Accent, en-USUniversity Students+3
DHH has gone completely off the rails...C1
1

DHH has gone completely off the rails...

American Accent, en-USUniversity Students+2
DNS Explained in 100 SecondsC1
2

DNS Explained in 100 Seconds

American Accent, en-USUniversity Students+2
Anthropic begged the world to stop AI… then shipped thisB2
3

Anthropic begged the world to stop AI… then shipped this

American Accent, en-USUniversity Students+2
One man just liberated Fable... and now it’s illegalB2
2

One man just liberated Fable... and now it’s illegal

American Accent, en-USUniversity Students+2
7 Things No Programmer Ever Wants to HearB2
2

7 Things No Programmer Ever Wants to Hear

American Accent, en-USBusiness English+1
80% of programmers are NOT happy… why?B2
1

80% of programmers are NOT happy… why?

American Accent, en-USUniversity Students+2
The most interesting hack in history just got weirder...C1
4

The most interesting hack in history just got weirder...

American Accent, en-USUniversity Students+2
Raspberry Pi Explained in 100 SecondsB2
4

Raspberry Pi Explained in 100 Seconds

American Accent, en-USUniversity Students+2
Did Google just kickstart the intelligence explosion?C1
2

Did Google just kickstart the intelligence explosion?

American Accent, en-USUniversity Students+3
Tragic mistake... Anthropic leaks Claude’s source codeC1
3

Tragic mistake... Anthropic leaks Claude’s source code

American Accent, en-USUniversity Students+2
100+ Docker Concepts you Need to KnowC1

100+ Docker Concepts you Need to Know

American Accent, en-USUniversity Students+2
When open-sourcing your code goes wrong...C1
7

When open-sourcing your code goes wrong...

American Accent, en-USUniversity Students+2
Docker in 100 SecondsC1
29

Docker in 100 Seconds

American Accent, en-USUniversity Students+2
The mystery is solved... and the answer is 40x cheaper than ClaudeC1
3

The mystery is solved... and the answer is 40x cheaper than Claude

American Accent, en-USUniversity Students+2
An ex-OpenAI researcher just deleted language from the LLM...B2
8

An ex-OpenAI researcher just deleted language from the LLM...

American Accent, en-USUniversity Students+2
25 crazy software bugs explainedB2
9

25 crazy software bugs explained

American Accent, en-USUniversity Students+2
✨ Vídeo recomendado
PyTorch in 100 SecondsC1
2
33 frases

PyTorch in 100 Seconds

American Accent, en-USUniversity Students+2
The Engineering that Runs the Digital World 🛠️⚙️💻 How do CPUs Work?C1
1
40 frases

The Engineering that Runs the Digital World 🛠️⚙️💻 How do CPUs Work?

University StudentsBusiness English+1
NVIDIA GTC 2025 - Keynote Highlights by NVIDIA CEO Jensen HuangC1
3
188 frases

NVIDIA GTC 2025 - Keynote Highlights by NVIDIA CEO Jensen Huang

University StudentsBusiness English+1
Unreal Engine 5.8 - Experimental MCP Server Support Walkthrough | State of Unreal 2026C1
1
197 frases

Unreal Engine 5.8 - Experimental MCP Server Support Walkthrough | State of Unreal 2026

University StudentsBusiness English+1
Google’s AI Course for Beginners (in 10 minutes)!C1
105 frases

Google’s AI Course for Beginners (in 10 minutes)!

American Accent, en-USUniversity Students+2
My Puzzle Robot is 200x Faster Than a HumanC1
6
249 frases

My Puzzle Robot is 200x Faster Than a Human

American Accent, en-USUniversity Students+2
🔥 Java Multithreading Rapid Fire | Can You Answer These Interview Questions? #shortsC1
3
26 frases

🔥 Java Multithreading Rapid Fire | Can You Answer These Interview Questions? #shorts

University StudentsBusiness English+2
Master n8n Triggers in 20 Minutes – Automate AnythingC1
1
333 frases

Master n8n Triggers in 20 Minutes – Automate Anything

University StudentsBusiness English+1
Amazon Athena explained in 120 seconds💡C1
15
29 frases

Amazon Athena explained in 120 seconds💡

University StudentsBusiness English+1
The best (and worst) setup upgrades.B2
1
131 frases

The best (and worst) setup upgrades.

University StudentsBusiness English+1
Rapid Engineering With AIC1
1
12 frases

Rapid Engineering With AI

University StudentsBusiness English+1

O que você vai aprender

Com este vídeo, você desenvolve três habilidades práticas: falar sobre tecnologia de forma clara, usar termos técnicos em inglês (como "GPU", "CUDA kernel" e "parallel computing") com confiança, e seguir uma explicação passo-a-passo em inglês, essencial para compreender conteúdos especializados. Além disso, treina a escuta de falas rápidas, comuns em aulas e tutoriais.

Ouça atentamente esses sons

A fala do narrador apresenta linking (ligação) entre palavras, como em "compute unified" (pronunciado como "computeunified") e "data scientists" (como "datascientists"). Há também reduções, como "it's" para "its" (em "it's free to attend") e "you've" para "youv" (em "you've got over 2 million pixels"). Observe como "GPU" é pronunciado com acento na primeira sílaba ("jee-pee-yoo") e "CPU" como "see-pee-yoo".

Fale como um nativo

Para imitar o ritmo do narrador, destaque palavras-chave como "revolutionized", "parallel", "neural networks" e "optimizing" com mais força. A fala é rápida, mas mantém pausas após frases longas, como "The Graphics Processing Unit, or GPU, is historically used for what the name implies, to compute graphics." Use essas pausas para respirar e manter a fluidez. Pratique o shadowing (imitação imediata) com trechos curtos, como a explicação sobre kernels: "You write a function, called a CUDA kernel, that runs on the GPU." Isso ajuda a melhorar a pronúncia em inglês e a sincronizar-se com a entonação nativa. Sites de shadowing, como o shadowspeaks, são excelentes para treinar essa técnica, pois permitem repetir frases até alcançar a naturalidade desejada. Lembre-se: a chave é a prática constante, focando na ligação de palavras e no destaque certo das sílabas.

O que é a Técnica de Shadowing?

Shadowing é uma técnica de aprendizado de idiomas com base científica, originalmente desenvolvida para o treinamento de intérpretes profissionais. O método é simples, mas poderoso: você ouve áudio em inglês nativo e repete imediatamente em voz alta — como uma sombra seguindo o falante com 1-2 segundos de atraso. Pesquisas mostram melhora significativa na precisão da pronúncia, entonação, ritmo, sons conectados, compreensão auditiva e fluência na fala.

Técnica de shadowing: leia o guia completo passo a passo →