Back in March, we announced that all App Engine apps hosted on appspot.com would be reachable over IPv6, from ISPs enrolled in the Google over IPv6 program.
Today, we're adding the same service to all domains hosted by ghs.google.com. If your App Engine app runs on your own domain name using Google Apps, then be sure to review our earlier blog post, which discusses the caveats of parsing client IP addresses, and how to use the SixXS Gateway to test your site over IPv6.
Currently, the Google over IPv6 program presents AAAA records only to ISPs within our whitelist. This helps ensure that our services remain reliable, by reducing exposure to clients with broken IPv6 connectivity. As the maturity of IPv6 deployment improves, we expect this whitelist to grow, eventually disappearing as it encompasses the entire Internet.
That said, we know that some of you are fans of IPv6 who would prefer to skip the whitelist, and offer dual-stack service to everyone immediately. To make your domain bypass the whitelist, just change your CNAME pointer from ghs.google.com to ghs46.google.com. Please note that, while ghs46 will cause your App Engine site to receive a larger fraction of its traffic over IPv6, our studies show that it may make your site slow or unreachable for up to 0.1% of clients across the Internet, so use this alternate domain with care.
As before, if you experience any problems with IPv6 serving and App Engine, please report it in the App Engine issue tracker or post about it in the App Engine discussion group.
Posted by Paul + the App Engine Traffic Team
If you have been running an App Engine application over the last few months, you will have noticed that performance of the Datastore has drastically improved. In fact, it quickly improved after the blog post we wrote in May acknowledging the performance issues. We chose to wait until this point to ensure that we could maintain the performance and quality of service. Thanks to a huge effort from the Datastore engineering team and our Production team, the metrics over the past month have made us confident that we have returned to a level of performance that we are proud to offer. With that in mind, we are now going to re-enable Datastore CPU usage billing.
On Tuesday, August 3rd, we will re-enable charging for Datastore CPU usage.
This does not mean that we are stopping work on Datastore performance and reliability-- quite the opposite, in fact. The issue has forced the team to step back and prioritize our long term roadmap for the Datastore. While we’re happy with our performance now, there are some very large changes to the Datastore in the pipeline that will continue to improve the App Engine platform. These include features such as a more highly replicated Datastore (now part of the roadmap) and improvements to the index building pipeline.
Thank you all for your patience and continued use of App Engine during this time. We’re excited to continue to serve our apps’ continued growth.
At Google I/O we announced the Mapper API. Built completely on top of public App Engine APIs today, this API is only the first component of App Engine’s MapReduce toolkit, but can be extremely useful on its own.
The Mapper API can already be of use to many developers who would otherwise need to build their own tool for doing large scale data manipulation. In addition to taking care of the distribution of these jobs over task queues, it provides the ability to store state, batch datastore writes via mutation pools, and ships with an easy to use administrative interface for job management, all optimized for the constraints of App Engine’s dynamic serving environment. Some examples of the types of operations that work with minimal configuration with this tool:
When you’re ready to jump in and start using the tool, head over to the project homepage on Google Code. You’ll want to check out the “Getting Started” page for the language you’re using:
- Fred, Mike, Ikai, Nick + the App Engine team
Fresh from our exciting news from Google I/O last quarter (launch of Google App Engine for Business and new product developments), we now look ahead to the mid-summer months. Since I/O, we've already appeared at a number of events including the Google DevFest Australia and Philippines, PyCon AU, and the Malaysia Open Source Conference, with many more coming this month. There are several well-known large conferences (and many smaller ones too) that our team members will be attending and/or speaking at, so hopefully we'll get to meet you at one of them!
In addition to the ones listed above, Google will be hosting more DevFests in Asia and South America later this year. Keep an eye out for specific dates and locations on the Google DevFest home page. There are also 5 larger global Google Developer Day events in the fall. We would love to meet with App Engine developers at a DevFest, Developer Day, or any of the other conferences above!
See you around the world!
Posted by Wesley Chun, Google App Engine Team
We have two scheduled upcoming maintenance periods for the Datastore:
As part of our on going Datastore reliability work, we'd like to inform developers that during the period between the two maintenance events listed above, we are expecting that Datastore performance will be impacted and applications will see higher read/write latencies. This maintenance and the higher latency in between are an unfortunate, but necessary step in our long-term plan towards completing the work to address the datastore latency issues (as described in our recent blog post). We recognize the inconvenience to applications and their users caused by this. For this reason, we continue to offer free datastore CPU usage until we are confident we have cleared these issues.
You can read more about the currently scheduled maintenance periods on the notification to the Downtime Notify list. As a reminder, be sure to sign up with the Downtime Notify list for alerts and updates on scheduled (or unplanned) maintenance, and follow @app_engine on Twitter for ongoing announcements.