Pratique du Shadowing: Mind-bending new programming language for GPUs just dropped... - Apprendre l'anglais à l'oral avec la vidéo
Choisissez un mode d'entraînement
Chaque mode ouvre une salle d'entraînement en plein écran.
60 phrases
Si les phrases sont trop courtes ou trop longues, cliquez sur Edit pour les ajuster.
1
Yesterday, the clouds opened up, and a weird new programming language came down to Earth, with a promise of parallelism for all thou who writeth code.
2
This is big if true, because parallel computing is a superpower.
3
It allows a programmer to take a problem that could be solved in a week, and instead solve it in seven days using seven different computers.
4
Unfortunately, running code in parallel is like conducting a symphony.
5
One wrong note, and the entire thing becomes a total disaster.
6
But luckily, Bend offers hope by making a bold promise.
7
Everything that can run in parallel, will run in parallel.
8
You don't need to know anything about CUDA, blocks, locks, mutexes, or regexes to write algorithms that take advantage of all 24 of your CPU cores,
9
or even all 16 ,000 of your GPU cores.
10
You just write some high -level
11
Python -looking code, and the rest is magic.
12
It is May 17th, 2024, and you're watching the Code Report.
13
When you write code in a language like
14
Python, your code runs on a single thread.
15
That means only one thing can happen at a time.
16
It's like going to a KFC with only one employee who takes the order, cleans the toilets, and cooks the food in that order.
17
Now, on a modern CPU, you might have a clock cycle around 4 gigahertz, and if it's handling one instruction per cycle,
18
you're only able to perform 4 billion instructions per second.
19
Now, if 4 Gips is not enough, you can modify your Python code to take advantage of multiple threads,
20
but it adds a lot of complexity to your code.
21
And there's all kinds of gotchas, like race conditions, deadlocks, thread starvation, and may even lead to conflicts with demons.
22
Even if you do manage to get it working, you might find that your CPU just doesn't have enough juice at
23
which point you look into using the thousands of CUDA cores on your GPU.
24
But now you'll need to write some C++ code and likely blow your leg off in the process.
25
Well, what if there is a language that just knew how to run things in parallel by default?
26
That's the promise of Bend.
27
Imagine we have a computation that adds two completely random numbers together.
28
In Python, the interpreter is going to convert this into bytecode, and then eventually run it on the Python virtual machine.
29
Pretty simple, but in Bend, things are a little more complex.
30
The elements of the computation are structured into a graph, which are called interaction combinators.
31
You can think of it as a big network of all the computations that need to be done.
32
When two nodes run into each other, the computation progresses by following a simple set of rules
33
that rewrite the computation in a way that can be done in parallel.
34
It continues this pattern until all computations are done, then merges the result back into whatever expression was returned from the function.
35
This concept of interaction combinators goes all the way back to the 1990s, and is implemented in a runtime called the higher -order virtual machine.
36
HVM is not meant to be used directly, and that's why they built
37
Bend, a high -level language, to interface with it.
38
And the language itself is implemented in Rust.
39
Its syntax is very similar to Python, and we can write a hello world by defining a main function that returns a string.
40
Now to execute this code, we can pull up the terminal and use the ben run command.
41
By default, this is going to use the Rust interpreter, which will execute it sequentially, just like any other boring language.
42
But now here's where things get interesting.
43
Imagine we have an algorithm that needs to count a bunch of numbers and then add them together.
44
The first thing that might blow your mind is that bend does not have loops.
45
Like, we can't just do a for loop like we would in Python.
46
Instead, bend has something entirely different called a fold that works like a search and replace for data types.
47
And any algorithm that requires a loop can be replaced with a fold.
48
Basically, a fold allows you to consume recursive data types in parallel, like a list or a tree.
49
But first, we need to construct a recursive data type.
50
And for that, we have the bend keyword, which is like the opposite of fold.
51
Now, if that's a little too mind -bending, maybe check out my back catalog for recursion in 100 seconds.
52
But now let's see what this looks like from a performance standpoint.
53
When I try to run this algorithm on a single thread, it takes forever, like 10 minutes or more.
54
However, I can run the same code without any modification whatsoever with the bend run C command.
55
When I do that, it's now utilizing all 24 threads on my CPU.
56
And now it only takes about 30 seconds to run the computation.
57
That's a huge improvement, but I think we can still do better.
58
Because I'm a baller, I have an NVIDIA RTX 4090, and once again, I can run this code without any modification on CUDA with bendrun -cu.
59
And now this code only takes one and a half seconds to run, and I'll just go ahead and drop the mic right there.
60
This has been the Code Report, thanks for watching, and I will see you in the next one.
Télécharger l'application
Tout pour parler couramment
Prononciation IAChaque phrase notée
Phonétique (IPA)Maîtrisez chaque son
VocabulaireEnrichissez votre lexique
Jeu de vocabulaireApprenez en jouant
Même chaîne
C11
The last real programmers are dying out...
Daily ConversationUniversity StudentsWorking Adults
C11
PostgreSQL in 100 Seconds
Daily ConversationWorking Adults
C12
Did Google just dethrone OpenAI and Anthropic?
Daily ConversationUniversity StudentsWorking Adults
C13
A $6.3 billion open-weight model just got embarrassed by the French...
Daily ConversationUniversity StudentsWorking Adults
C13
Every app on Earth is now open source... whether it likes it or not
Daily ConversationWorking Adults
C13
OpenAI’s lead engineer was secretly a spy...
Daily ConversationUniversity StudentsWorking Adults
B26
Intel chips can’t possibly be this bad… 100% crash rate?
American AccentBusiness EnglishUniversity StudentsWorking Adults
C12
Godot in 100 Seconds
American AccentBusiness EnglishUniversity StudentsWorking Adults
B21
Grok 4 pushes humanity closer to AGI… but there’s a problem
American AccentBusiness EnglishUniversity StudentsWorking Adults
C13
Array Map in 100 Seconds
American AccentBusiness EnglishUniversity StudentsWorking Adults
C130
Docker in 100 Seconds
American AccentBusiness EnglishUniversity StudentsWorking Adults
C17
The most interesting "hack" in history...
American AccentBusiness EnglishUniversity StudentsWorking Adults
B23
OpenAI’s new browser feels familiar…
American AccentBusiness EnglishUniversity StudentsWorking Adults
C112
10 weird OSS projects you need right now...
American AccentBusiness EnglishUniversity StudentsWorking Adults
C216
NestJS in 100 Seconds
American AccentBusiness EnglishUniversity StudentsWorking Adults
B23
Firefox and Tor hit with 9.8 critical level exploit
American AccentBusiness EnglishUniversity StudentsWorking Adults
B23
5 open source tools that replaced my $320/mo AI stack...
American AccentBusiness EnglishUniversity StudentsWorking Adults
C12
C++ in 100 Seconds
American AccentBusiness EnglishUniversity StudentsWorking Adults
C11
DHH has gone completely off the rails...
American AccentBusiness EnglishUniversity StudentsWorking Adults
C12
Rust in 100 Seconds
American AccentBusiness EnglishUniversity StudentsWorking Adults
FOR YOU
Vidéo recommandée
C11
462 phrases
Expensive RTX 5090 for LLMs? NO. Use This Instead. (Tesla V100 SXM2 + Z8 G4, #RACERRRZ)
Daily ConversationUniversity StudentsWorking Adults
B21
64 phrases
What is Python? Why Python is So Popular?
Daily ConversationUniversity StudentsWorking Adults
C11
108 phrases
How I Learned Procedural Generation
Daily ConversationUniversity StudentsWorking Adults
C132 phrases
Linus Torvalds Explains What Makes Linux So Fast
Daily ConversationUniversity StudentsWorking Adults
C115
29 phrases
Amazon Athena explained in 120 seconds💡
Daily ConversationUniversity StudentsWorking Adults
C11
68 phrases
GitHub Copilot now controls your command line...
Daily ConversationUniversity StudentsWorking Adults
C117 phrases
Very excited to share that I now have a column in New Scientist
Daily ConversationUniversity StudentsWorking Adults
C185 phrases
We Can Now Simulate a Human Brain, Scientists Show
Daily ConversationUniversity StudentsWorking Adults
C11
40 phrases
The Engineering that Runs the Digital World 🛠️⚙️💻 How do CPUs Work?
Daily ConversationUniversity StudentsWorking Adults
C11
207 phrases
Go’s Big Mistake
Daily ConversationUniversity StudentsWorking Adults
Vocabulaire et conseils d’expression pour cette leçon
Cette leçon d’expression orale s’appuie sur la vidéo « Mind-bending new programming language for GPUs just dropped.. ». Les mots qui reviennent le plus souvent : code, bend, computation, Python, parallel. Cette vidéo contient 60 phrases et 904 mots à répéter en shadowing. La partie parlée dure 4:00. Le locuteur parle vite, environ 226 mots par minute : attendez-vous à des liaisons et à des sons réduits. Seuls 77 % des mots font partie des 3 000 mots les plus courants en anglais, le vocabulaire est donc exigeant.
Vocabulaire clé de cette vidéo
15 mots de la vidéo à retenir, avec leur prononciation et leur sens :
| Mot | Prononciation | Sens |
|---|---|---|
| bend verbe | /bɛnd/ | courber, tordre |
| parallel adjectif | /ˈpæ.ɹəˌlɛl/ | parallèle |
| thread nom | /ˈθɹɛd/ | fil |
| fold verbe | /ˈfəʊld/ | plier |
| algorithm nom | /ˈælɡəɹɪðm̩/ | algorithme |
| loop nom | /luːp/ | boucle |
| solve verbe | /sɒlv/ | résoudre, régler |
| virtual adjectif | /ˈvɜɹt͡ʃuəl/ | virtuel |
| default nom | /dɪˈfɔlt/ | défaut, non-respect |
| interaction nom | /ˌɪn.təˈɹæk.ʃən/ | interaction |
| implement verbe | /ˈɪmpləmənt/ | appliquer, exécuter |
| instruction nom | /ɪnˈstɹʌkʃən/ | instruction |
| execute verbe | /ˈɛksɪˌkjuːt/ | exécuter, mettre à mort |
| cloud nom | /ˈklaʊ̯d/ | nuage |
| element nom | /ˈɛləmənt/ | élément |
Les verbes à particule que vous entendrez
| Mot | Prononciation | Sens |
|---|---|---|
| check out verbe | mater, regarder | |
| go ahead verbe | /ˌɡoʊ əˈhɛd/ | vas-y, allez-y |
| open up verbe | ouvrir | |
| pull up verbe | /ˌpʊl ˈʌp/ | soulever |
La grammaire de cette vidéo
Les structures que le locuteur utilise le plus, avec les mots exacts de la vidéo :
| Structure | Dans la vidéo |
|---|---|
| Voix passive be + participe passé — l’accent est mis sur ce qui arrive, pas sur qui le fait | could be solved · are structured · be done |
| Propositions relatives who / which + proposition — une précision sur une personne ou une chose | graph, which are · interpreter, which will · keyword, which is |
Prononciation à surveiller
Le locuteur utilise 8 contractions et formes réduites, comme you're, can't, doesn't. Prononcez-les sous leur forme courte, telles que vous les entendez.
- Les sons « th »: thread /ˈθɹɛd/, algorithm /ˈælɡəɹɪðm̩/
- Les sons « sh » et « zh »: interaction /ˌɪn.təˈɹæk.ʃən/, instruction /ɪnˈstɹʌkʃən/, modification /ˌmɑdɪfɪˈkeɪʃən/, starvation /stɑɹˈveɪʃən/, computation /ˌkɑmpjʊˈteɪʃn̩/
- Mots longs — placez bien l’accent: interaction /ˌɪn.təˈɹæk.ʃən/, modification /ˌmɑdɪfɪˈkeɪʃən/, interpreter /ɪnˈtɝ.pɹɪ.tɚ/, complexity /kəmˈplɛk.sɪ.ti/, whatsoever /ˌ(h)wʌtsoʊˈɛvɚ/
Les sons difficiles pour les francophones :
- /tʃ/ et /dʒ/ — à ne pas adoucir en « ch » et « j »: virtual /ˈvɜɹt͡ʃuəl/, juice /d͡ʒuːs/, merge /mɝd͡ʒ/, gotcha /ˈɡɑ.tʃə/, sequentially /səˈkwɛntʃəli/
- /r/ anglais — langue recourbée, sans frotter la gorge: parallel /ˈpæ.ɹəˌlɛl/, thread /ˈθɹɛd/, algorithm /ˈælɡəɹɪðm̩/, virtual /ˈvɜɹt͡ʃuəl/, interaction /ˌɪn.təˈɹæk.ʃən/
Comment s’entraîner avec cette vidéo
- Écoutez la vidéo en entier une fois sans parler et notez les mots que vous ne connaissez pas.
- Commencez à la vitesse 0,75×, répétez phrase par phrase, puis revenez à la vitesse normale quand cela devient facile.
- Enregistrez-vous et comparez avec l’original, en faisant attention à des mots comme bend, parallel, thread.
Qu'est-ce que la technique du Shadowing ?
Le Shadowing est une technique d'apprentissage des langues fondée sur la science, développée à l'origine pour la formation des interprètes professionnels. Le principe est simple mais puissant : vous écoutez de l'anglais natif et le répétez immédiatement à voix haute — comme une ombre suivant le locuteur avec un décalage de 1 à 2 secondes. Les recherches montrent une amélioration significative de la précision de la prononciation, de l'intonation, du rythme, des liaisons, de la compréhension orale et de la fluidité.