Back in June 2015, Google commissioned a whitepaper discussing AWS vs. Google Cloud pricing. As you might imagine, the authors of the whitepaper answered the question is Google Cloud cheaper than AWS with a resounding “yes”. However, a lot has changed since June 2015.
Whereas Amazon has pursued a policy of regular price cuts, Google snatched the carpet from under the market leader´s feet with the launch of custom instances in November 2015. The AWS vs. GCP pricing wars continued in 2016 with AWS launching Reserved Instances, and Google releasing Committed Use and Sustained Use Discounts - effectively informing business how they could optimize GCP costs.
Then, in September 2017, Amazon announced it was to offer per-second billing for EC2 instances, EBS volumes, and several other services. Within two weeks Google responded by extending its existing per-second billing program to Compute Engine instances running on both Linux and Windows servers. The sequence looks set to continue, especially with other cloud service providers such as Alibaba looking to extend into the U.S. and Europe.
Because of factors such as custom GCE instances, AWS Reserved Instances and ongoing price reductions, it is difficult to conduct a meaningful AWS vs. Google Cloud pricing comparison. Where similarities can be found, AWS tends to be less expensive on a CPU/Hour basis for most workloads, while GCP is the cheaper option for compute intensive workloads.
This might surprise a few people - particularly the authors of the 2015 whitepaper - however, the comparison was conducted on an hour-by-hour comparison ignoring GCE´s Sustained Use Discounts and Inferred Instances. Once these factors are taken into account, Google Cloud is cheaper than AWS in most use cases - it just takes a lot of working out.
Naturally businesses will be concerned about obtaining the best value for money when comparing AWS vs. Google Cloud pricing, but decisions should not be made on AWS vs. GCP cost comparisons alone. Factors such as flexibility, ease of use, reliability and performance should also enter calculations, in which case Google Cloud Platform wins hands down.
However, AWS has an advantage in cloud features and points of presence around the globe. AWS offers many more regions and availability zones than GCE, which can affect the speed at which it is possible to access and use resources deployed in the cloud. This might be an important consideration for organizations with multinational operations - particularly outside the U.S. and Europe.
Returning to AWS vs. GCP pricing, one important point to be aware of is vendor lock in. In cloud computing terms, this is the scenario in which a business commits to a level of service in order to receive a discount. With AWS, businesses can increase the discount by paying upfront for one or three years, or by partially paying up-front and paying the remainder monthly. GCP does not offer an advanced payment option, but will contract you to a certain payment every month.
One of the biggest issues with vendor lock-in is that businesses do not benefit from subsequent price reductions once they have committed to a level of service. Furthermore, if a business commits to one service provider, and another introduces features that would substantially benefit its operations, there is no way to change. This is one of the reasons why businesses consider hybrid or multi-cloud options.
With an increasing number of businesses moving towards hybrid and multi-cloud environments (which also have the advantage of improved disaster recovery planning), one has to ask whether a Google Cloud vs. AWS price comparison is still relevant. Certainly, with the challenges involved in creating a like-for-like AWS vs. GCP cost comparison, many compute, memory, storage and network types fall outside comparison parameters - creating a false indication of AWS vs. Google Cloud pricing.
Furthermore, with quarter-by-quarter triple-digit growth in container deployments, price comparisons matching instances against instances bear no likeness to actual cloud costs. In our free-to-download eBook “Are You Ready for Containers”, we report Gartner´s forecast that “by 2018 more than 50% of new workloads will be deployed into containers in at least one stage of the application lifecycle.”
Therefore, with the advantages of hybrid and multi-cloud environments, and with negligible price differences between container deployments, a Google Cloud vs. AWS price comparison is not an effective way of calculating which the better option is - if the objective is to use one option exclusively. Instead businesses should look at optimizing cloud costs by implementing software suitable for multi-cloud cloud monitoring and cost optimization at container level.
Source: Triple digit growth