YT YouTube2Text

YouTube2Text → Transcripts → AWS Containers & Serverless Explained | ECS, EKS, Fargate & Lambda | AWS CLF-C02 Day 9

AWS Containers & Serverless Explained | ECS, EKS, Fargate & Lambda | AWS CLF-C02 Day 9 — Transcript

by Pawan Joshi · 4,304 words · 668 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00Welcome back everyone to day nine of our
0:01AWS builder challenge. Yesterday we
0:03launched our very first EC2 instance. We
0:05connected to a Linux server. We explored
0:07compute, storage and networking as well
0:10as security. But today we are going to
0:12ask an important question. What if I
0:15don't want to manage servers at all?
0:18Think about it. As a developer, I don't
0:21enjoy patching the servers, caring about
0:24operating system updates, or do you like
0:26wake up excited to configure the
0:28infrastructure first? Probably not. Most
0:32developers want to focus on building the
0:33application, writing the logic, writing
0:35the code, solving problems, creating new
0:37features. This desire is what led to two
0:40major revolutions in cloud computing.
0:42Containers and serverless computing.
0:46Today we will learn how cloud platforms
0:48evolve from virtual machines to
0:50containers and eventually serverless.
0:53And by the end of this session you will
0:55create your first AWS Lambda function.
0:57You will write code, you will deploy it,
0:59you will execute it and see as
1:02automatically handles the infrastructure
1:04behind the scenes and that's pretty
1:05amazing when you think about it. So
1:08let's begin. Let's quickly take a look
1:10at our today's agenda, today's road map.
1:13First we will understand the evolution
1:15of cloud computing. We will compare
1:18virtual machines, v containers, v
1:20serverless.
1:22Then we will explore AWS container
1:24services. You will hear several new
1:26names today like ECS, EKS, ECR, fog
1:30rate. By the end of this session, you
1:33will understand exactly what each
1:35service does and when to use it. After
1:37that, we will dive into AWS Lambda. AWS
1:41Lambda is one of the most popular
1:43services. It is uh it's also one of the
1:46most heavily tested serverless services
1:48in the cloud practitioner exam. It's
1:51also one of the most heavily tested
1:53serverless services in the cloud
1:54practitioner exam. We will discuss how
1:57lambda works, what is its pricing, what
2:00are the lambda limits and also we will
2:03discuss about some common lambda exam
2:05scenarios. Then we'll learn to decide
2:08like when to choose lambda, when to
2:10choose EC2 or far grid because AWS often
2:13ask architecture based question in the
2:15exams and finally we will create our
2:18first AWS lambda function together. We
2:20will write code, deploy it and test it.
2:24Then we will also review the logs in the
2:27cloudatch. We will understand what
2:28cloudatch is. By the end of this video,
2:30you will have deployed code to the cloud
2:32without launching a single server. So
2:34this slide is one of the most important
2:37conceptual slides in the today's lesson.
2:39To understand Lambda, to understand
2:41Fargate and to understand the containers
2:43as well, you first need to understand
2:45how cloud computing actually evolved. So
2:47let's start with virtual machines.
2:50Imagine you own an entire house. In this
2:54house, you have walls, plumbing,
2:56electricity, furniture, maintenance,
2:58everything. And this is similar to
3:01virtual machines with EC2 provides the
3:03physical hardware. EC2 provides physical
3:06hardware and you are responsible for a
3:09lot of things like in your home you are
3:11responsible for all these things.
3:12Plumbing, electricity, furniture. In a
3:14similar way in EC2 you choose the
3:18operating system, you install the
3:19software, you pass the operating system
3:21and updates is managed by you. security
3:24scaling everything you have maximum
3:27flexibility that is you have full
3:29control but with full control comes
3:33maximum responsibility
3:36okay think back to the studies EC2
3:39instance we launched the server
3:40connected to Linux updated packages by
3:42ourselves then installed software that's
3:44exactly VM management looks like now
3:48let's move one step further and let's
3:49discuss about container
3:52imagine
3:53Instead of owning house, now you own an
3:56apartment.
3:58Now you don't manage the building. You
4:00don't manage elevator, security guard,
4:02parking lot, nothing. The building owner
4:05handles all of that. You focus only on
4:07your apartment. Containers work in a
4:10very similar way. Instead of packaging
4:13an entire operating system, containers
4:15package application, libraries, its
4:18dependencies and also its configuration,
4:22everything the application needs to run.
4:24Containers are much lighter. They are
4:28much lighter in compared to the virtual
4:29machines and they start faster and they
4:32use fewer resources and it makes
4:35application deployment much easier. One
4:38of the most common container
4:39technologies is Docker. Some of you
4:43might have used the Docker as well. So
4:45the key idea is build once and run
4:47anywhere. If it works on your laptop, it
4:50should work in the cloud too. Now let's
4:52move to the serverless. What do this
4:55serverless means? Imagine staying in a
4:58hotel room. You don't manage the
5:00building. You don't manage elevators. No
5:03maintenance. You don't manage cleaning
5:05as well. What you do? You simply
5:08check in. You simply check in into your
5:11room and use the room and leave and you
5:13pay only the for the time you actually
5:16use it. And that's serverless
5:18technology. Example include lambda. With
5:21AWS Lambda, you don't have to manage
5:23servers, operating system, configure,
5:25scaling, nothing. You want you don't
5:27have to worry about anything
5:28infrastructure as well. What you do is
5:30just give the code to the AWS. AWS will
5:33handle everything else. This is why
5:35lambda is so powerful and it's also why
5:37modern cloud applications increasingly
5:40rely on serverless architecture also the
5:42platform that you are using currently
5:46that might be using serverless
5:48technology. So let me ask you a few
5:51question if it is said that which model
5:54gives the most control what would you
5:56say virtual machine? Yes. And if it is
6:00asked which model requires the least
6:02infrastructure management
6:05the answer will be serverless. And
6:08talking about
6:10model that packages application and
6:12dependencies together that is container.
6:15Now we have understanded containers
6:17conceptually. Let's see how AWS provide
6:19container services. So today we are not
6:22deploying containers and that's more of
6:24the solution architects or developer
6:26associate topic. But I want you to see
6:28where these services actually live
6:30inside the AWS and what they look like.
6:32So let me show you ECS service. What I
6:35can do is go to the search bar and
6:37search ECS. You will find this elastic
6:40container service. Let me open this. So
6:43now what I can do is go to the clusters
6:45and we do create cluster for that. Uh as
6:48you can see you can click on cluster and
6:50name the cluster setting forget only
6:52forget managed whatever you need. You
6:55can set up monitoring encryption as well
6:57and then click on create and your
6:58cluster is done. So we are not going to
7:01create in this session but we might see
7:04in further sessions okay but we don't
7:06need this. Now let me show you another
7:09service that we are going to read about
7:11that is EKS.
7:13From the name only these services are
7:15clear. If said ECS elastic container
7:19service
7:21EKS elastic Kubernetes service you can
7:26see EKS is Kubernetes on AWS. For cloud
7:29practitioner the most important thing to
7:30remember is simple. If the question
7:32mentions Kubernetes
7:34think EKS and that's enough. Let me show
7:37you ECS as well. Uh we have done ECS.
7:40Let me show you ECR that is elastic
7:43container registry. Containers are
7:46packaged as images and those images need
7:49somewhere to live and that's exactly
7:51what ECR provides. You can create a
7:53repository, set up the repository name
7:56and create the repository.
7:59And there is also another service we
8:01talked about far. What do that means? If
8:05you go to ECS, we saw that. Let me show
8:09you. Go to cluster and I will click on
8:12create cluster. Now you can see you have
8:15option for far. Fargate only that is
8:17serverless. You don't need to think
8:19about creating or managing servers.
8:20Fargate and manage instances means that
8:23is a bit managed. Manage instances means
8:26Amazon ECS will manage the patching and
8:28scaling on your behalf while giving you
8:30the configurability of about the types
8:33of instances. Great for more advanced
8:34workloads. Yeah, you can select whatever
8:37you need
8:39and this is the key idea. ECS manage the
8:41container and Farget provides the
8:43serverless compute for those containers.
8:46Now we have seen AWS container services
8:49on the AWS console itself. Now we can
8:52read and easily understand what these
8:55services actually are. So AWS provides
8:59several services for running containers
9:01on it. At first these names can sound
9:03confusing like EKS, ECS, ECR what these
9:07are. But by the end of this slide you
9:09will know exactly what is service does.
9:11You don't need to worry about that. So
9:13let us start with ECS. ECS. ECS stands
9:16for elastic container service. Think of
9:19ECS like an AWS native container
9:22management platform. So its job is
9:24simple. Run and manage Docker
9:26containers. Imagine your application
9:28consist of five containers. instead of
9:31manually starting and monitoring each
9:33one. So, ECS manages them for you. It
9:36can start containers,
9:38it can stop containers, it can restart
9:41the failed containers and it can scale
9:43containers as well, distribute
9:44workloads. And the most important thing
9:45to remember here is ECS is AWS own
9:50container orchestration service and
9:53because AWS built it, it's generally
9:55easier to learn than Kubernetes. For
9:57many organization, ECS is the simplest
9:59way to run containers on AWS. Let's talk
10:02about EKS. Has anyone heard of
10:04Kubernetes before? You don't need to
10:07know Kubernetes for cloud practitioner,
10:09but you should know what EKS is. EKS
10:12stands for elastic Kubernetes service.
10:15Okay. So remember the term C here, K
10:18here. C means container, K means
10:21Kubernetes.
10:23So AWS didn't create Kubernetes.
10:25Kubernetes is an open-source container
10:27orchestration platform originally
10:28created by Google. But yeah, many large
10:31organization use Kubernetes because it
10:33works across multiple cloud providers.
10:35AWS created EKS so customer could run
10:38Kubernetes cluster without managing all
10:40the underlying complexity. Think of it
10:43like ECS equal to AWS native solution
10:46and EKS is the Kubernetes on AWS. Got
10:50it? Exam tip what I want to give is if a
10:54question specifically mentions
10:55Kubernetes the answer is usually EKS.
10:58Now let's discuss one of my favorite
11:00services that is Fargate. Imagine you
11:03want to run a container or you want to
11:05run containers but you don't want to
11:07manage the EC2 instances. You don't want
11:10to manage operating system. You don't
11:12want to manage patching. You don't want
11:15to manage the server maintenance just
11:17continuous. That's exactly what Fargate
11:19provides. Fargate is a serverless
11:22compute for containers. You provide the
11:24container and the AWS provide the
11:27infrastructure. And here's very
11:28important point. Fargate is not a
11:31container orchestration platform.
11:33Remember that Fargate is a compute
11:35engine. It works together with ECS or
11:39EKS. Think of ECS and EKS are the
11:42managers and Fargate as a worker doing
11:45the actual compute. Okay, let me tell
11:47this again. ECS and EKS are the
11:52container orchestration platform, but
11:54Fargate is a compute engine. Think of
11:57ECS and EKS as the managers and think of
11:59Fargate as the worker doing actual
12:01compute. The distinction appears
12:02frequently in certification exam. So you
12:04must remember the difference between
12:05them. The last one is ECR, elastic
12:09container registry. Containers are
12:11stored as images. But where do those
12:13images live? That's where ECR comes in.
12:16ECR stands for elastic container
12:18registry. Think of ECR as AWS private
12:22docker image registry. Secure storage
12:25for container images. If docker hub is a
12:28public image library, ECR is your
12:30private AWS version. Developers push
12:32container images to ECR and then ECS or
12:36EKS. It pulls the images from them and
12:40runs them. Easy.
12:43Now if the question is asked that the
12:46company wants to run containers without
12:48managing EC2 servers, what will be the
12:50answer? Think carefully.
12:53The answer is Fargate.
12:55Why? Because this is the compute engine
12:58for containers. Remember that. And if
13:01the company specifically requires
13:03Kubernetes, then you use EKS. Where are
13:08those container images stored? they are
13:10stored in registry that is ECR that is
13:13elastic container registry. Okay. So you
13:16must have understood that what is ECS,
13:19EKS, Fargate, ECR. Now let's move to the
13:23one of the most important AWS services
13:25ever created that is AWS Lambda. If EC2
13:30changed how we think about servers,
13:32Lambda changed how we think about
13:34applications. Before Lambda existed,
13:37developers typically launch servers
13:38before deploying code. With Lambda, AWS
13:42flipped that model. See, instead of
13:44managing the servers, instead of
13:46managing the servers,
13:48you simply upload the code. That's it.
13:52AWS handles everything. Infrastructure,
13:56operating system, scaling, availability,
14:00patching, monitoring, everything. You
14:03just focus on the code. That's why
14:05lambda is called serverless computing.
14:07Now, here's a key question. When does
14:10the lambda run? The answer is when an
14:12event occurs. That's not the case with
14:15the EC2. Okay? EC2 runs continuously
14:19until you stop it. That's why it is
14:21called eventdriven architecture.
14:24Whenever there is event, then the lambda
14:27runs. So, let's take a look at some
14:29examples. Suppose someone requires to
14:32upload files to S3. The upload event
14:36automatically triggers the lambda. So
14:38upload event will automatically triggers
14:40the lambda and lambda processes the
14:41file. There is no requirement for
14:43server.
14:44Suppose user submit a form to the
14:47website then the API gateway receives a
14:50request. API gateway triggers the lambda
14:52and lambda process the data. Again there
14:54is no required for ser server. Similar
14:57goes for these all Dynamob streams,
15:00cloudatch events, S, SQS message, SNS
15:02notification. You might hear these terms
15:06in the further sessions as well. AWS
15:08provides one of the most generous free
15:10tiers for Lambda. The first 1 million
15:13requests per month are free. So every
15:15month you get this much request and
15:18these are a lot. And unlike many other
15:20older free tier benefits, this always
15:23free offer does not expire. that makes
15:25lambda perfect for learning. So now let
15:28us see key limits in pricing. This is
15:30one of the most tested lambda fact like
15:33maximum execution time 15 minutes. Let
15:36me repeat that a single lambda execution
15:39cannot run longer than 15 minutes. If
15:42your task need 20 minutes, lambda is not
15:44the right choice. You would likely need
15:46EC2, ECS or farate. Student often get
15:50this wrong in the exam. Remember 15
15:52minutes is the maximum per single
15:55function execution. So actually what is
15:57there in lambda? We do get a function.
15:59Let's say I have this link. Okay, I
16:02click on this link. The lambda gets
16:05triggered and I get the result. But if
16:08this action takes more than 15 minutes
16:12then no it it's not the right choice.
16:15Okay. Lambda is maximum of 15 minutes
16:17timeout. Another thing lambda scaling.
16:20One of the coolest thing about Lambda is
16:22scaling. Suppose one person triggers
16:23your function. AWS run one execution.
16:26Suppose 10,000
16:29person trigger the same function
16:31simultaneously. AWS automatically
16:34scales. There is no need for ASG. That
16:36is called autoscaling groups. There is
16:38no capacity planning required. No server
16:40provisioning. It simply works. So what
16:42happens is for each user a copy of code
16:46is created and it is served to that
16:48user. So you don't need to worry about
16:51infrastructure hardware when you are
16:53learning about lambda or when you are op
16:55when you are working with lambda.
16:57So whenever you see eventdriven
17:01or serverless or automatic scaling paper
17:07request okay no server management think
17:11of lambda and this is almost every time.
17:14So now let's bring everything together
17:17because this is exactly how AWS asks
17:19question in the certification exam. They
17:21won't ask like what is lambda, what is
17:23ECS. Instead they will give you a
17:25scenario and you will choose the correct
17:27service.
17:28Now let us see when to use which
17:30service. Lambda is used for shorter task
17:33less than 15 minutes and event driven
17:36architecture and some they are used for
17:38unpredictable traffic spikes and the
17:41priority is pay per use. only pay for
17:44what you use. EC2 when you need full OS
17:47control when you are thinking to manage
17:48each and everything yourself in this the
17:51customer has the maximum responsibility.
17:54Talking about ECS on far
17:57it is used for Docker container
17:59workloads microservices architecture and
18:02no server management desired. Talking
18:05about EKS, if you have already invested
18:07in Kubernetes and you need complex
18:11container orchestration features, then
18:13you use EKS. EKS means elastic
18:15Kubernetes service. EC2 means elastic
18:18compute cloud. ECS means elastic compute
18:21service and EKS means elastic kubernetes
18:25service. Remember that. So in exam
18:27pattern, let's say if ask serverless
18:30plus event-driven plus autoscaling that
18:31is lambda. So in exam the questions
18:36won't be like what is lambda instead
18:38they will give you a scenario and you
18:39will choose the correct service. So
18:40let's take the scenario a company
18:42process images whenever the user uploads
18:44file to S3. Okay traffic traffic is
18:49unpredictable they want zero server
18:51management. So what will be the answer?
18:53The answer will be lambda. Yes. So like
18:57whenever the file is uploaded lambda
19:00function is triggered and with that
19:03function we process the image that is
19:06going to be uploaded in the S3. So this
19:09is textbook lambda use case. A company
19:13wants to run a large application that
19:15requires full operating system control.
19:17What would be the answer? Amazon EC2.
19:21Let's say a company has Docker
19:22containers and want to avoid managing
19:24servers. Which service we use? We will
19:27use Fargate. And a company let's say
19:30already have invested in Kubernetes and
19:32across multiple environments.
19:35Across multiple environments, what will
19:36be the answer? EKS because Kubernetes is
19:39the requirement. There is also one more
19:41thing that is called AppRunner.
19:44Let me tell you simply about that. Let's
19:47say a startup wants the simplest way to
19:49deploy a web application.
19:51The answer is AppRunner. What does it
19:54do? We provide the source code. Click on
19:56deploy and done. Minimal infrastructure
19:58management. So remember always identify
20:01the keywords in the exam like
20:02serverless, lambda, kubernetes, EKS,
20:05container without services, farget or
20:07full control if said that is EC2. So
20:09this approach solves most architecture
20:12question. So now it's time to stop
20:14talking about lambda and actually build
20:16something. Let's create our first
20:17serverless application. So you will
20:19understand how they actually work. So
20:21now let's see. We are going to explore
20:23lambda. Now let me go to lambdas. I will
20:27simply search for lambda over here. lb
20:30bda. Okay. Open this. And now see I do
20:35have a button called create a function.
20:38You can click on create a function
20:41and then author from scratch. Let me
20:43give it a function name. Let's say say
20:45hello. So it will say hello to me. Okay.
20:49and then click on create and let me see
20:53okay I will choose the runtime as Python
20:55first as I I'm going to use Python for
20:58writing the code and do I need anything
21:00other okay so I don't need anything
21:03other I will click on create function
21:06so here we didn't create a server we
21:08didn't choose RAM we didn't configure
21:10storage we simply told here's my
21:13function that's all that's the power of
21:17serverless computing Now I have a
21:20function here that is lambda function.
21:22We have code over here and it says hello
21:24to AWS SPG. Okay. So you can see
21:30function name over here. You can see you
21:32can see test button to test the event
21:35monitor configuration and many others
21:39also. You can find this function ARN
21:42just like S3 bucket and IM users. Notice
21:45this ARN. Lambda functions have unique
21:47AWS identifiers. What was ARN? ARN was
21:51Amazon resource names. Okay. So what you
21:55can do is let's write our own code. Def
22:00let me make my own function named LMBDL
22:03lambda
22:06handler. Let me give it this event and
22:09context.
22:11And let me return something from here.
22:13Okay. return what I will return is let
22:17me return with the status code status
22:20code
22:21200
22:23and there will be a message along with
22:26this that is hello from AWS lambda.
22:30So once we are done let us try if it
22:33actually works or not. So what we did is
22:36we simply u added a function in here in
22:40lambda we uploaded our code or we wrote
22:43our code. Now AWS automatically handle
22:45the servers, scaling, availability,
22:47operating system, infrastructure. We
22:48don't need to manage anything at all.
22:50Now let us test it. Let's create a test
22:52event. So we can test if our function is
22:54working or not. I will go to create. I
22:57will go to test then create new event.
23:00And now set some event name like my
23:04let's say first test. My first test.
23:07Okay. Then you will go to event JSON and
23:11you can simply write name AWS
23:16builder. Got it? Let us test this. Okay.
23:20As you can see status code status code
23:23is 200 body hello from lambda.
23:26So you can find each and every
23:27information um like how much time did it
23:30take to run build duration how much time
23:34for how much time we are built max
23:36memory used resources configured each
23:38and everything in real world application
23:41events often come from S3 uploads API
23:44gateway event bridge Dynamob so today we
23:47have triggered it manually so
23:48congratulations you have just executed
23:51one code inside AWS without launching a
23:53single server AWS just created the
23:56infrastructure. What you did was
23:58executed your code, returned the result
24:00and then cleaned everything up
24:01automatically.
24:05Now let us see eventdriven example. So
24:08now let me slightly change the code and
24:10show you how eventdriven things work. Um
24:14so let me add name over here. Name equal
24:18to event dot
24:22get name world. Okay. And inside this uh
24:26hello what I will do is refer to this
24:30and yeah got it. All right. So I guess
24:34it is correct name. All right. So now I
24:38can deploy the function. See lambda
24:40functions receive events and events are
24:42simply the pieces of information sent to
24:43the function. So we are sending the name
24:47over here. So let us test if it is
24:49working or not. Click on test and then
24:52hello AWS builder.
24:54Why it showed hello AWS builder is that
24:58because in the test we have already
25:01written name as AWS builder. That is why
25:04it showed me AWS builder. Let me test it
25:06again. Hello Ram. Now let us test this.
25:10You can see hello Ram. All right. So
25:13it's working. So now you can see our
25:16Lambda function is working fine. Now let
25:19us explore the cloudatch logs. Now let
25:22us see the cloudatch logs. Simply you
25:24can search for cloudatch or yes I have
25:28this over here
25:30and then what I will do is go to
25:33dashboards
25:34and I will go to automatic dashboard and
25:36search for lambda. I can simply see my
25:39logs over here. Every lambda execution
25:42um automatically generate log. We didn't
25:44install a login server or we didn't
25:45install or configure storage. AWS
25:48automatically integrates lambda with the
25:50cloudatch. You can see how many
25:53invocations were done and you can see
25:56the functions was updated when its
25:58memory size, code size. So you can find
26:02event duration, memory, use, request ID,
26:05each and everything from here. Now we
26:07will simply go back to lambda and delete
26:10the function we have just created. Why
26:12we are deleting this? Do it charges even
26:15if we are not using it. But we said as
26:17we will use we will pay but we are not
26:20invoking this function then we won't be
26:22charged. It's true but it's a good
26:24practice to clean up the resources that
26:27we have uh created and we are if we are
26:30not using them. So did we create a
26:32server? No. Did we configure
26:34autoscaling? No. What triggered our
26:37function? a test event and where we can
26:40view event logs, execution logs in the
26:44cloudatch logs.
26:46Remember the maximum time for this AWS
26:49Lambda execution is 15 minutes only.
26:53Okay. So there's a bonus thing for you
26:55all. Let me show you how we can trigger
26:58the event. Let's say I have say hello
27:01only. I will select NodeJS and click on
27:03create function. Let me show you the
27:06available triggers. Actually, in like
27:08real world architectures, Lambda is
27:09actually triggered automatically by
27:10events occurring across the AWS
27:13services. We don't manually trigger
27:15every time. That's not the approach. Uh
27:18so I guess the function is created. No.
27:20Yeah. Okay. Creating the function say
27:22hello. So now if you see the
27:24configuration tab, you can find
27:27different ways to trigger. As you can
27:30see trigger you can add trigger you can
27:33select the source like even your
27:35function needs to be executed. So you
27:37can search for S3 if any event in S3
27:40occurs or you can select API gateway.
27:44Okay let's see I'm going to use API
27:46gateway. You can create new API or you
27:48use an or even use an existing API. Let
27:51me see. You can create HTTP API webocket
27:53rest API. For web application you may
27:56use these rest API HTTP or websocket
27:58API. uh we actually use serverless
28:00technology for that serverless framework
28:03for that. Um we will also understand
28:05this in some of the projects in the
28:07upcoming videos but let's but simply
28:10this is what it is. Okay, day nine
28:14summary. So what we have done, we have
28:16seen what do we mean by VNS, container
28:18and serverless. We saw about ECS, EKS,
28:21Fargate things and lambda essentials
28:25like what is eventdriven architecture,
28:27what is the limit for that and how many
28:29requests are free and stuff. So what I
28:31want you to do is today practice 10
28:34questions on lambda. You might find them
28:36on the node itself. And tomorrow we will
28:39explore one of the most versatile and
28:41widely used storage service in AWS that
28:43is Amazon S3.

About this transcript

This page contains the full transcript of AWS Containers & Serverless Explained | ECS, EKS, Fargate & Lambda | AWS CLF-C02 Day 9 by Pawan Joshi, generated from the public captions YouTube serves with the video. The transcript has 4,304 words across 668 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.