Google Compute Engine Release Announcement

[Google Compute Engine Team]

Greetings Google Cloud Platform users,

Today, we announced new features for the Google Cloud Platform which included a highly requested feature for Compute Engine: Layer 3 Load Balancing. We're also releasing a new version of gcutil, which includes Load Balancing commands and resource types, as well as a change to the default behavior of the addinstance command. Finally, we are announcing a new open source GitHub project for the utilities used in Compute Engine Images. Read on for more information about each of these exciting announcements.

Layer 3 Load Balancing

From our blog announcement:

We’ve now added Layer 3 Load Balancing to Compute Engine, which delivers Google-scale throughput and fault tolerance to manage Internet applications. Load balancing is critical in any highly scalable system, allowing you to automatically and intelligently route traffic across a collection of servers.

With the load balancing service, you can:

• Load-balance ingress network TCP/UDP traffic over a specific set of Compute Engine virtual machines (VMs) within the same region

• Ensure that only healthy VMs are used to serve Internet requests through the use of HTTP-based health checks

• Easily handle spikes in load without pre-warming

• Configure the load balancer via command line interface (CLI) and a programmatic RESTful API

This initial release provides Layer 3 support and we’ll continue to expand its capabilities on a regular basis.  We’re pleased to offer the load balancing feature at no cost through the end of 2013, after which we’ll charge the rates outlined on our pricing page.

Load Balancing and gcutil v1.8.3

In order to use Load Balancing with gcutil you will need to upgrade to version v1.8.3.

NOTE: There is a new prompt that has been added to gcutil's addinstance command, so we recommend reading the documentation if you use that command in any of your scripts.

After upgrading gcutil, you can start using the Load Balancing service by creating Forwarding Rules and Target Pools to distribute your traffic across a set of specified instances. You can also add health checking to your load-balanced instances using the new Http Health Checking resource. We've briefly described these new load balancing resources below.

Target Pool Resources: A Target Pool contains a list of associated Instances that should receive load-balanced traffic. When a Forwarding Rule resource (defined below) receives traffic, it will send the traffic to the specified Target Pool and the traffic is then distributed across Instances in that Target Pool.

To create a Target Pool resource:

gcutil --project=${PROJECT_ID} addtargetpool www-pool --region=${REGION} \

 --instances=${ZONE}/www1,${ZONE}/www2

Forwarding Rule Resources: A Forwarding Rule matches and directs traffic to the associated Target Pool, where it is it distributed across Instances associated with the Target Pool.  A forwarding rule will match traffic based on the IP address and protocol:port that the traffic is originating from.

To create a forwarding rule resource:

gcutil addforwardingrule www-rule --region=${REGION} --port_range=80 --target=www-pool

To learn more about how to use these resources together, review the Load Balancing quick start guide. For more information about each of these new resources, review the Load Balancing documentation.

Image utilities are now on GitHub

We are happy to announce a new GitHub project to share the source code for the utilities that are used by Google Compute Engine Images. You can use the source code to help you more easily build and maintain public and custom Images. For more information about how to use these utilities to create Images, please see our image bundling documentation.

The GitHub project includes the following utilities:

• Image Bundle - Tool that creates an image file out of a disk attached to a GCE VM.

• Google Startup Scripts - Scripts and configuration files that setup a Linux-based image to work smoothly with GCE.

• Google Daemon - A service that manages user accounts, maintains ssh login keys, and syncs public endpoint IP addresses.

That's about it for now; we hope you are as excited about these announcements as we are.

Thanks, and happy computing!

-- Paul Rashidi, on behalf of the Google Cloud Platform