A truly amazing and detailed video. Has really enhanced my understanding of global accelerator. I have watched several other videos of yours to grasp the details on other topics such as RDS, Aurora, DynamoDb, Route 53 routing policies, Kinesis, etc. Thanks for all these videos.
Mar 29, 2021 - AWS Global Accelerator now supports traffic through the new AWS edge ... and redirects traffic to healthy endpoints in less than 30 seconds.
Where the global accelerator physically lies so its so fast? On the closest edge location and reduce hops to 1? Or AWS has some direct connections to a lot ISPs?
@@Pythoholic So for example, if we have created global accelerator in Acount A and we created 2 ALB in Account B and Account C. So can we add those two ALB endpoint into global accelerator in Account A?
Yes, it is possible to add Application Load Balancers (ALBs) in different AWS accounts to a Global Accelerator created in another AWS account. To achieve this, you would need to follow these general steps: Create a Global Accelerator in Account A and note down the DNS name and the static IP address associated with it. In Account B and Account C, create the ALBs and configure them to serve traffic to the desired targets. Create an Endpoint Group for each of the ALBs in the same AWS Region as the ALB in each account. Configure the Endpoint Group to use the ALB's DNS name as the endpoint and select the desired health check configuration. Add the Endpoint Groups to the Listener for the Global Accelerator in Account A. You can do this by specifying the listener port, protocol, and endpoint group ARN for each of the endpoint groups. Test the configuration by sending traffic to the Global Accelerator DNS name or IP address.
Thanks for amazing content , I just have 2 question about the topic. First , why we need or this service provides us 2 static ip address since we're gonna use dns name on the client side. what is the meaning of those static ip addresses ? Second, in the video you said client doesn't cache dns name to ip address mapping when using global acceleration dns name. Why it doesn't cache it and how it happens ?
1. **Why Two Static IP Addresses?** - **Redundancy and Fault Tolerance**: Two static IP addresses provide redundancy. If one IP experiences issues, the other can continue to serve traffic, ensuring high availability. - **Global Traffic Distribution**: These IPs are anycast from multiple AWS edge locations. This means that a user's request is automatically routed to the nearest edge location for lower latency. - **Ease of Whitelisting**: Static IPs simplify firewall configurations for clients who need to whitelist IPs. Changing IPs can be problematic in such scenarios. - **DNS Simplification**: While DNS names are used for human-friendly access, the underlying static IPs remain constant, providing a stable endpoint. 2. **Why Doesn't the Client Cache DNS Name to IP Address Mapping?** - **Dynamic Routing Optimization**: AWS Global Accelerator continuously monitors the health of your application endpoints and routes traffic to the nearest healthy endpoint. If a client caches the DNS mapping, they might continue to route traffic to an endpoint that has become unhealthy or suboptimal in terms of latency. - **TTL Settings**: The DNS name provided by services like AWS Global Accelerator typically has a very low Time To Live (TTL), which discourages DNS resolvers from caching the address. This ensures that the resolving service frequently checks for the most optimal endpoint. Essentially, the non-caching of DNS names ensures that traffic is always routed in the most efficient manner, taking into account the current network conditions and the health of your application endpoints. This dynamic rerouting is crucial for maintaining performance and availability, especially in a globally distributed environment.
@@Pythoholic When the infra is placed close to the customer, what is the need for a global accelerator? This is quite confusing because I feel that everything serves the same purpose. Or am I understanding it wrongly?
Its like in local zone we are building the whole infra for that specific purpose and it will have its resources closer to the people who need it. global accelerators are helpful to reduce the latency for users who are far from the hosted location. but local zones are specifically setup closer to meet that purpose.
Hi, Sir first of all I need to thank you for making such quality content free on RUclips. Sir, I have one Personal question for you. I am working in a non Technical industry, I like to switch my career to AWS. I don't have knowledge of coding. Whether I can make it to AWS Domain with just learning through online courses and certification. What should I really need to do to get into the AWS domain? I will be so thankful to you if you reply to my comment
Hi Mohan, Its good that you are planning to move into AWS, My advice will be to working on your resume as per the current market trends, Working on small projects independently by taking online sources so that you can add them to your resume. If you wish to move forward with a certification you should do that, Probably starting off with SAAC02 or any specialization certs that you want to learn, there is a list of specializations available. Mostly companies look out for individual with added knowledge like cloud and cicd, along with the programming knowledge. So make sure you are applying for the right job. So prepare your resume well, with enough credentials to stand out. Starting afresh might be bit tough but you need to be persistent. And keep giving interviews, to know the expectations. But to answer your question, Yes you can. But i would advice you to keep trying this along with your current job that you have. That would be safe.
Hi Sir, you brought a smile to my face, I am very happy to receive a reply from you, If I got any further doubts can I contact you back. Is there any alternate way to contact you? I promise you I will not disturb you unnecessarily.
What's the diff b/w the Global accelerator and Cloudfront for dynamic content(caching disabled) ? Let's say i have dynamic web application deployed in EC2 and it has global user. Which service should i use to reduce the network latency?
I think u have to try and relate it in a way that what ur requirement is. Cloudfront with edge locations most suited for accessing static content faster across the globe. But global accelerator does a lot more than just http.traffic. you are trying to connect to a endpoint using the Aws network and inturn which reduces ur latency. You need to think on aspects such as price, dependent configuration, what kind of access point u are looking for. I hope it helps.
@@Pythoholic Hi, Thanks for the reply :) . Doesn't cloud front support Dynamic content. As far as i understood, it support dynamic content by disabling the cache in edge location which inturn use the global network through which it reduce the network latency. Please correct me if my understanding is wrong.
Yeah it does, but if I had to do that and I am hosting the application I would just create a global accelerator and attach the load balancer endpoint and that's it. I can share the URL and I am set. Cloudfront is an added overhead in that sense. Is what I feel. It uses the global network and the edge locations to find the shortest path so it makes it easier for me to achieve the low latency I desire. If I disable caching for cloudfront, that would defeat the purpose. Because my goal is to just just reduce the.latency of my users using my application hosted in EC2. If it has nothing to do with the static content then I am better off without it.
It would depend on if your WordPress site is hosted on AWS or another platform. if it's on another platform using CloudFront as CDN would be a good option to make it faster aws.amazon.com/blogs/networking-and-content-delivery/accelerating-wordpress-with-cloudfront-using-the-aws-for-wordpress-plugin/
If we have applications in one location. May be using GA will be a costly affair. But yes it will improve the performance which is already in a good state.
In AWS Global Accelerator, you are charged for each accelerator that is provisioned and the amount of traffic in the dominant direction that flows through the accelerator
This makes latency WORSE, I've had first hand experience of Global accelerator playing VALORANT which is hosted by AWS. My ping was worse by 70%, ended up using a VPN to AVOID global accelerator BS, ping went down from 60ms ping 35ms immediately. DO NOT USE THIS until it's more widespread with CDN in EVERY major city.
A truly amazing and detailed video. Has really enhanced my understanding of global accelerator. I have watched several other videos of yours to grasp the details on other topics such as RDS, Aurora, DynamoDb, Route 53 routing policies, Kinesis, etc. Thanks for all these videos.
Till now best video on Global acclerator on RUclips
@@basiddivekar9366 aye haye kya bola biru. Chai to pila
I am really amazed to see the efforts and quality put on to make this video. Wonderful presentation.
Excellent explanation esp in differentiating cloudfront from GA! Thanks
Great video!! Amazing quality and breakdown
Nice tutorials, thank you. What software do you use for the animations?
Very thorough. Thanks for the elaborate video.
Very detailed explanation 💯🙌
Thais is a very good explanation of global accelerator!
Nice one which explains the concepts crisp and clear. Thanks for your efforts!!
at 8:41 it should be 30ms instead of 30sec I hope :(
amazingly detailed as always btw, thanks :)
Mar 29, 2021 - AWS Global Accelerator now supports traffic through the new AWS edge ... and redirects traffic to healthy endpoints in less than 30 seconds.
Where the global accelerator physically lies so its so fast? On the closest edge location and reduce hops to 1? Or AWS has some direct connections to a lot ISPs?
Ok so its like CDN with same edge locations but for TCP/UDP messages instead of resources. Last part of video is most important
could you help explain diff b/w Route53 latency routing and global accelerator?
are endpoint group different from VPC endpoints ?
Very detailed video and explained very well.
I had one doubt - can we add other account ALB endpoint to another account global accelerator.
Thanks for the feedback* if you could please provide some details for the query*
@@Pythoholic
So for example, if we have created global accelerator in Acount A and we created 2 ALB in Account B and Account C.
So can we add those two ALB endpoint into global accelerator in Account A?
Yes, it is possible to add Application Load Balancers (ALBs) in different AWS accounts to a Global Accelerator created in another AWS account.
To achieve this, you would need to follow these general steps:
Create a Global Accelerator in Account A and note down the DNS name and the static IP address associated with it.
In Account B and Account C, create the ALBs and configure them to serve traffic to the desired targets.
Create an Endpoint Group for each of the ALBs in the same AWS Region as the ALB in each account. Configure the Endpoint Group to use the ALB's DNS name as the endpoint and select the desired health check configuration.
Add the Endpoint Groups to the Listener for the Global Accelerator in Account A. You can do this by specifying the listener port, protocol, and endpoint group ARN for each of the endpoint groups.
Test the configuration by sending traffic to the Global Accelerator DNS name or IP address.
@@Pythoholic Thank You for the detailed answer. Will check it.
Best video with Clear explanation. Can I use Global accelerator for https application.?
It depends on the hosting type -- this is just going to help route better.
Thanks for amazing content , I just have 2 question about the topic. First , why we need or this service provides us 2 static ip address since we're gonna use dns name on the client side. what is the meaning of those static ip addresses ? Second, in the video you said client doesn't cache dns name to ip address mapping when using global acceleration dns name. Why it doesn't cache it and how it happens ?
1. **Why Two Static IP Addresses?**
- **Redundancy and Fault Tolerance**: Two static IP addresses provide redundancy. If one IP experiences issues, the other can continue to serve traffic, ensuring high availability.
- **Global Traffic Distribution**: These IPs are anycast from multiple AWS edge locations. This means that a user's request is automatically routed to the nearest edge location for lower latency.
- **Ease of Whitelisting**: Static IPs simplify firewall configurations for clients who need to whitelist IPs. Changing IPs can be problematic in such scenarios.
- **DNS Simplification**: While DNS names are used for human-friendly access, the underlying static IPs remain constant, providing a stable endpoint.
2. **Why Doesn't the Client Cache DNS Name to IP Address Mapping?**
- **Dynamic Routing Optimization**: AWS Global Accelerator continuously monitors the health of your application endpoints and routes traffic to the nearest healthy endpoint. If a client caches the DNS mapping, they might continue to route traffic to an endpoint that has become unhealthy or suboptimal in terms of latency.
- **TTL Settings**: The DNS name provided by services like AWS Global Accelerator typically has a very low Time To Live (TTL), which discourages DNS resolvers from caching the address. This ensures that the resolving service frequently checks for the most optimal endpoint.
Essentially, the non-caching of DNS names ensures that traffic is always routed in the most efficient manner, taking into account the current network conditions and the health of your application endpoints. This dynamic rerouting is crucial for maintaining performance and availability, especially in a globally distributed environment.
Thanks for comprehensive reply @@Pythoholic
Very well explained as usual. Thanks
Great video, to the point. thanks..
Great Video Thanks for this informative session.
Excellent tutorials with nice diagrams!!!
Nice one with details very good
Thank you for the detailed explanation. I have a quick question on global accelerator. Does it also work for local zones?
Well it should, but local zones purpose is to place the infrastructure close to the customer. So not sure if this is a right fit for that requirement.
@@Pythoholic When the infra is placed close to the customer, what is the need for a global accelerator? This is quite confusing because I feel that everything serves the same purpose. Or am I understanding it wrongly?
Its like in local zone we are building the whole infra for that specific purpose and it will have its resources closer to the people who need it. global accelerators are helpful to reduce the latency for users who are far from the hosted location.
but local zones are specifically setup closer to meet that purpose.
Hi, Sir first of all I need to thank you for making such quality content free on RUclips. Sir, I have one Personal question for you. I am working in a non Technical industry, I like to switch my career to AWS. I don't have knowledge of coding. Whether I can make it to AWS Domain with just learning through online courses and certification. What should I really need to do to get into the AWS domain? I will be so thankful to you if you reply to my comment
Hi Mohan, Its good that you are planning to move into AWS, My advice will be to working on your resume as per the current market trends, Working on small projects independently by taking online sources so that you can add them to your resume. If you wish to move forward with a certification you should do that, Probably starting off with SAAC02 or any specialization certs that you want to learn, there is a list of specializations available. Mostly companies look out for individual with added knowledge like cloud and cicd, along with the programming knowledge. So make sure you are applying for the right job. So prepare your resume well, with enough credentials to stand out. Starting afresh might be bit tough but you need to be persistent. And keep giving interviews, to know the expectations. But to answer your question, Yes you can. But i would advice you to keep trying this along with your current job that you have. That would be safe.
Hi Sir, you brought a smile to my face, I am very happy to receive a reply from you, If I got any further doubts can I contact you back. Is there any alternate way to contact you? I promise you I will not disturb you unnecessarily.
No worries..You can surely send me a mail with ur queries.
You are the best out there
thanks vinay
a new follower right here ;)
What's the diff b/w the Global accelerator and Cloudfront for dynamic content(caching disabled) ? Let's say i have dynamic web application deployed in EC2 and it has global user. Which service should i use to reduce the network latency?
I think u have to try and relate it in a way that what ur requirement is.
Cloudfront with edge locations most suited for accessing static content faster across the globe.
But global accelerator does a lot more than just http.traffic. you are trying to connect to a endpoint using the Aws network and inturn which reduces ur latency.
You need to think on aspects such as price, dependent configuration, what kind of access point u are looking for.
I hope it helps.
@@Pythoholic Hi, Thanks for the reply :) . Doesn't cloud front support Dynamic content. As far as i understood, it support dynamic content by disabling the cache in edge location which inturn use the global network through which it reduce the network latency. Please correct me if my understanding is wrong.
Yeah it does, but if I had to do that and I am hosting the application I would just create a global accelerator and attach the load balancer endpoint and that's it.
I can share the URL and I am set.
Cloudfront is an added overhead in that sense. Is what I feel. It uses the global network and the edge locations to find the shortest path so it makes it easier for me to achieve the low latency I desire.
If I disable caching for cloudfront, that would defeat the purpose.
Because my goal is to just just reduce the.latency of my users using my application hosted in EC2. If it has nothing to do with the static content then I am better off without it.
Can we link a wordpress site to global accelerator with https?
It would depend on if your WordPress site is hosted on AWS or another platform. if it's on another platform using CloudFront as CDN would be a good option to make it faster aws.amazon.com/blogs/networking-and-content-delivery/accelerating-wordpress-with-cloudfront-using-the-aws-for-wordpress-plugin/
Most apt explanation
very good video, nice explained
a big thankyou sirji
thanks 🙏
if my application is available in one location instance will GA help in improving latency?
If we have applications in one location. May be using GA will be a costly affair. But yes it will improve the performance which is already in a good state.
Is AWS Global Accelerator comes under free tier?
In AWS Global Accelerator, you are charged for each accelerator that is provisioned and the amount of traffic in the dominant direction that flows through the accelerator
Very Informative
Thanks lot
This makes latency WORSE, I've had first hand experience of Global accelerator playing VALORANT which is hosted by AWS. My ping was worse by 70%, ended up using a VPN to AVOID global accelerator BS, ping went down from 60ms ping 35ms immediately. DO NOT USE THIS until it's more widespread with CDN in EVERY major city.
So true.
Cache content...not cage.
Hehe did i say that?? 😃😃
Cache in USA is pronounced as cash. Anyway, awesome content. Thank you so much.