article thumbnail

New Route 53 and ELB features: IPv6, Zone Apex, WRR and more.

All Things Distributed

New Route 53 and ELB features: IPv6, Zone Apex, WRR and more. Route53 support for WRR load balancing. Weighted Round Robin is a DNS based load-balancing scheme that allows you to select multiple IP address for a name to be resolved to and to assigns frequencies ("weights") to how often the name should resolve to each address.

IPv6 114
article thumbnail

Liveblog: IPv6 in the Cloud - Protocol and Service Overview

Scott Lowe

This is a liveblog of an AWS re:Invent 2017 breakout session titled “IPv6 in the Cloud: Protocol and Service Overview.” Global IPv6 adoption is currently around 22%, and is expected to hit 25% by the end of the year. T-Mobile, for example, now has 89% of their infrastructure running on IPv6.

IPv6 60
Insiders

Sign Up for our Newsletter

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Trending Sources

article thumbnail

Technology Short Take 153

Scott Lowe

This article contains some good information on IPv6 for those who are just starting to get more familiar with it, although toward the end it turns into a bit of an advertisement. However, it quickly runs out of steam, and you’ll need to progress to more advanced examples and documentation to continue your learning journey.

Linux 74
article thumbnail

Simplifying IT - Create Your Application with AWS CloudFormation.

All Things Distributed

There are several resources required: Elastic Load Balancers, EC2 instances, EBS volumes, SimpleDB domains and an RDS instance. A simple scenario is for example the ability to clearly identify production from staging and development environments. New Route 53 and ELB features: IPv6, Zone Apex, WRR and more.

article thumbnail

Amazon EC2 Cluster GPU Instances - All Things Distributed

All Things Distributed

For example, the most fundamental abstraction trade-off has always been latency versus throughput. The different stages were then load balanced across the available units. GPUs work best on problem sets that are ideally solved using massive fine-grained parallelism, using for example at least 5,000 - 10,000 threads.