We just launched AWS Global Accelerator! https://aws.amazon.com/blogs/aws/new-aws-global-accelerator-for-availability-and-performance/ … With it, you can use our global backbone to accelerate your applications. Now, allow me to nerd out and go deep on how some of the fault tolerance actually works ...
-
Show this thread
-
Global Accelerator uses BGP anycast. We give you an IP address that we announce from lots of locations on our backbone and "The Internet" figures out getting your users to the closest site.
2 replies 5 retweets 14 likesShow this thread -
Until now we've haven't offered anycast services, because fault tolerance is very very hard. We actually tried it for CloudFront years ago, but found it wasn't reliable or precise enough. So what's new?
1 reply 4 retweets 11 likesShow this thread -
The core challenge with anycast is that when a link dies, the networks connected to that link think that it's still where they should be sending packets. For a while at least ... up to minutes. This can cause congestion and black-holing ... which isn't, you know, good.
2 replies 2 retweets 9 likesShow this thread -
With Global Accelerator we give you what we've been doing for Amazon Route 53 for 8 years now ... striped, resilient, IPs. Here's what that means: you can have multiple static IPs .. and they are meaningfully redundant. We don't advertise these IPs on the same links!
2 replies 6 retweets 17 likesShow this thread -
So if there's a problem with a link, and downstream networks aren't caught up yet, only one IP should be affected. The other will be fine. You can still hardcode both IPs, they're both static and just for you, and ordinary client-level retries and resilience will take care of it.
1 reply 3 retweets 8 likesShow this thread -
This kind of "striping" for an anycast service is very very reliable, and much better plain anycast. I love seeing how our customer's priorities (availability in this case) end up being reflected in our products. It'll be awesome to see what applications customers build! FIN.
1 reply 2 retweets 13 likesShow this thread -
Replying to @colmmacc
So, do end users of my service still use global anycast to find the AWS edge and then the AWS edge uses this service to route to the 'best' (as determined by health measurements/settings) region?
1 reply 0 retweets 0 likes -
Replying to @Todd_Segal
Yes! You can create proximity-based, weighted active-active or failover configurations and the AWS Edge network and global backbone will figure out how to get the traffic there.
1 reply 0 retweets 0 likes -
Replying to @colmmacc
Nice, thanks for confirming! Do you have any services/features to route users to the 'best' edge location (and avoid peering problems, BGP hijacks, etc)?
1 reply 0 retweets 0 likes
We still have Amazon Route 53 Latency Based Routing. With that you can register multiple endpoints, tag them with an AWS region, and we figure out the best region to send users to; based on actual constantly-running Internet latency probes.
-
-
Replying to @colmmacc
so am I correct that: a. end users --> edge is optimized by latency based DNS updates to your anycast IPs b. edge --> origin is optimized by your new Global Accelerator service?
0 replies 0 retweets 0 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.