The cloud services business is a trillion dollar market. There was a time when Amazon (AWS) was dominating this market. Google launched own cloud service in 2011, and Microsoft unveiled its Azure cloud platform around the same time. Although the top spot for the most used cloud platform belongs to Amazon, Google and Azure are not far behind. Check out our previous post on AWS and GCP comparison.
Google and Azure are in close competition not only by dropping price, but also by launching several useful features. However, sometimes the two cloud platforms seem to trail each other so closely that the overall value delivery becomes indistinguishable. If the size and price of Amazon cloud are not for you, the next best choices will be limited to Google and Azure. But how should a user choose between the two cloud offerings?
Google has better Compute Engine
Google offers a powerful Compute Engine that has been available to all the users since 2013. This Engine allows users to operate virtual CPUs and categorize those into groups and regions. It also comes with useful features like virtual machine live migration, load balancing, better core availability, speedy persistent disks, and extended OS support. Azure doesn’t really have a comparable system. Azure did debut a compute service of its own about a year before Google did, but it has not been improved or made available to all the users like the Google engine has been since inception. Users can assign virtual machines with Azure using a separate service that requires getting predefined by the user, Microsoft, or a third party. Azure allows users to assign memory capacity and number of cores to a virtual machine. However, features stop there and Google clearly has a superior product in this aspect.
Google has a slightly better pricing
Both Google and Azure have a minute-based on-demand pricing structure. Both are quite flexible when compared to an hour-based service like the one Amazon offers. On demand here means that customers don’t need to pay an upfront cost or reserve usage in advance. You pay as per your usage. Google has a minimum of 10 minutes of usage and rounds up the total usage time to the number of minutes. Azure does pretty much the same thing.
Google, however, has upgraded the pricing structure to calculate sustained use. It means that the longer someone uses the service, the cheaper it would be. Google offers discounts for prolonged use of its service. This structure allows Google to have an excellent price advantage over Amazon. But for Azure, the service offers discounts for some short-term commitments or reserved usage. So, that gives Google a slight advantage here when it comes to pricing.
Azure offers hybrid cloud networking
Azure offers a virtual network that is similar to the popular Virtual Private Clouds offered by Amazon. These virtual networks allow users to group or categorize virtual machines as isolated networks on the platform. Users can then go on to define network topology, create route tables, set subnets, assign private IP addresses, or make network gateways. Essentially, this allows services to extend beyond a single data center premise on the public cloud. Google’s networking service, in comparison, is available on a single network. Users can define gateway addresses and ranges for all instances on this network. Users can enhance security by receiving public IP addresses and applying firewall rules to an instance. But Google doesn’t offer hybrid cloud services like Azure does in this regard.
Google offers fully supported storage and archiving
Google offers users ephemeral storage in which storage starts and gets subsequently deleted with an instance. They also offer persistent disk storage to hold files longer and Cloud Storage for objects. And all stored files are supported with Google’s additional applications throughout. For example, Google supports stored relational databases via the Cloud SQL service. Google’s most popular services like Big Table, Big Query, and Hadoop are fully supported with storage. They also offer archiving with Nearline, which is relatively cheap but has no recovery latency.
Azure, in comparison, offers only a “D drive” for temporary storage. Microsoft’s block storage option called Page Blobs are available on the platform as well. Azure does support NoSQL and relational databases. A limited number of services, like HDInsight and Windows Azure Table are available for big data. However, there is no archiving support available.
Google offers perks to indie developers
Google has a brand new service called ‘Always Free’, which comes with usage tier available for many of its products at no additional cost. While it doesn’t make a big difference for heavy users, Always Free offers excellent advantages to indie developers and cash-strapped start-ups that might want to test prototypes or set up private beta testing. Azure doesn’t have a comparable service. But, they allow users to add virtual networks with no additional cost.
The primary advantage that Google offers is that the services are expanding and improving at a steady pace. Google Cloud has come a long way since its debut about 6 years ago. In comparison, Azure, which was launched around the same time lags behind in certain key service areas.