Competition is good: Google Expands Cloud Platform


Google Logo

From the “Competition is Good” department, we have this just in from Google — reduced pricing, new features, new services, and expanded cloud service footprint around the globe:

 We’re constantly making updates to our Google Cloud Platform products—Google App Engine, Cloud Storage, Big Query, Compute Engine and others—based on user feedback and to improve the overall experience. For example, two weeks ago we introduced a major update to Google Cloud SQL providing faster performance, larger databases (100GB), an EU zone, and a no-cost trial. But, we know there is more to do. Today, we’re continuing to improve the platform with new storage and compute capabilities, significantly lower prices, and more European Datacenter support.

On the storage front, Google announced standard storage pricing reductions of 20% and introduced “Durable Reduced Availability” (DRA) storage, a new storage tier that trades off storage availability for an even lower price point. DRA looks to compete with Amazon’s reduced redundancy storage (RRS) that offers a similar tradeoff. The fundamental difference is that in return for lower pricing Google is offering reduced availability while AWS is offering reduced redundancy. DRA seems better when the data is critical, but you have flexibility for when you might want to retrieve it. In contrast, AWS RRS stores fewer copies of the data and so is better for data that is less critical.

Google also expanded its European data center support for App Engine, Cloud Storage, and Cloud SQL. Google Compute Engine will be available in Europe soon as well. This expanded data center support will be important for improving cloud-based application performance for European customers and for servicing European enterprises that might be subject to data sovereignty restrictions and otherwise prohibited from using Google’s USA-based services.

Google announced 36 new Compute Engine instance types and reduced pricing by about 5% on previously announced instance types. This puts continued pressure on AWS and other cloud service providers for both breadth of offering as well as pricing. Google now has a rich set of standard, high memory, and high CPU instance types to cater to different application requirements.

Finally, Google announced a snapshotting feature for persistent disks. Google says the new feature “…makes it simple to instantly create a backup of your disk, move it around Google datacenters, and use the snapshot to start up a new VM.” Depending on the limitations surrounding movement, this could be interesting. We assume “move it around” means between different Google data centers, which could provide an interesting capability for data backup and geographic VM migration, possibly for disaster recovery. Previously, Google had recommended this be done manually with rsync. Unfortunately, the announcement and the links provided don’t supply enough detail at this point.

As this announcement comes on the eve of AWS re:Invent, we can assume it was meant as a shot across the AWS bow, putting Amazon on notice that Google intents to compete heavily in cloud services going forward. At this point AWS has a large lead and many premier customers. I would expect customer announcements from Google to be quickly forthcoming.

Speak Your Mind

*