Know Your Cloud: Evaluating Amazon Web Services (AWS) and Google Cloud Platform

Jun 08, 2017
Share article

Although Amazon Web Services (AWS) still leads the public cloud market, the competition looks to be heating up. According to recent data [1], AWS currently holds over 40% of the market share — more than all other IaaS cloud services combined. But, Google’s Cloud Platform is gaining ground with growth in its segment, outpacing ad revenue growth [2] in Q1 2017.

But what made AWS the market leader? And how does Google Cloud Platform stack up? We’ve evaluated the two services in this comparison to help your organization decide which cloud service is best for your bottom line.

Amazon Web Services (AWS)

AWS has been in business since 2006, longer than any other provider. Based on Red Hat Enterprise Linux, it supports several Linux distributions, Windows servers 2003, 2008 and 2012, and Xen hypervisor and Docker containers. AWS has an unrivaled breadth of industry experience and an equally impressive array of complementary services and partners. Unsurprisingly, it’s the public cloud service of choice for the likes of Adobe, Netflix, and business messaging app Slack, among many other companies.

Where AWS Excels

AWS’ standout features are its high level of customizability and the sheer amount of options the service offers. Simply put, there isn’t much it can’t do. By adding elastic beanstalk configuration files to your source code, you can create a custom environment, configure it to include the resources you need, and even choose which software to run your application on. More to the point, whether you need a conventional cloud service for storage, virtual servers, or even a back-end for an IoT device, chances are AWS has a scalable solution.

AWS Snowball [3] can also handle petabyte data transfers securely in under one day via a lightweight, easy to install appliance. And if you need a database management system, AWS Aurora [4] is compatible with both MySQL and PostgreSQL, giving you unparalleled flexibility.

Where AWS Falls Short

The myriad options offered by AWS is also occasionally its downfall. Basic setup isn’t too complicated, but getting everything just right can involve a significant learning curve. Additionally, though AWS is often touted as one of the more affordable cloud services, many of its additional offerings have a separate pricing structure. As a result, overall cost can be difficult to estimate correctly.

Google Cloud Platform

Google Cloud Platform runs on Debian by default; however, it also supports 64-bit CentOS, Red Hat Enterprise Linux, SUSE, Windows 2008 and 2012, and Ubuntu Linux. For container support, firms can choose between Docker or CoreOS standards. Both are managed via Google’s management tool, Kubernetes.

In addition to solid computing, storage, database, and networking feature buckets, Google Cloud Platform also offers machine learning services and a serverless, fully integrated big data platform [5].

Where Google Cloud Platform Excels

Google Cloud Platform shines when it comes to big data and machine learning. Google’s edge in big data isn’t completely surprising, seeing as the company originally developed MapReduce [6]. That said, it really goes above and beyond by offering an extremely comprehensive range of solutions, from managed data warehousing (BigQuery [7]) to large scale exploration (Cloud Datalab [8]) and even genomic [9] data processing.

Similarly, Google Cloud Platform’s machine learning capabilities offer in-depth solutions, giving organizations access to the APIs of other Google apps. These range from well-known and publicly available services — such as Google Translate — to products they’ve built for in-house purposes, including Cloud Natural Language [10], which analyses unstructured text, and Cloud Vision [11], an image analytics tool.

Where Google Cloud Platform Falls Short

Despite its apparent strengths, Google Cloud Platform still has significantly fewer software options than AWS, making for a somewhat less flexible solution. Google has acquired some products in recent years to help fill some gaps, noticeably Stackdriver for monitoring and the real-time database platform Firebase. However, even with these acquisitions there are gaps between services offered by Google and AWS.

What’s more, while AWS offers 42 availability zones [12] in 16 regions worldwide, Google Cloud Platform currently only offers coverage in 20 zones [13] across 7 regions. This puts Google Cloud Platform at a significant disadvantage on two fronts. First, running a cloud server from the US — where Google has most of its availability — can be more expensive. Second, depending on your location, you may also run into latency issues, which could have a negative impact on user experience.

The Bottom Line

Because of its sheer size, comprehensiveness, and customizability, AWS is hard to beat. That said, Google Cloud Platform is a solid service, and it clearly has the edge when it comes to big data analysis and machine learning — AWS may have a hard time catching up on that front.