Sela. | Cloud Better.

Save your Cloud Costs by migrating to Google Cloud

Google of course is very well known to all of us. Search is the first thing that comes to our mind when we think of Google. But along with Search, a lot of other Google applications are also a part of our lives and we use them daily — Gmail, Android, Chrome, YouTube, Maps, Google Apps, etc.

By Yogesh Golande, Head of Engineering, Sela

 

Google of course is very well known to all of us. Search is the first thing that comes to our mind when we think of Google. But along with Search, a lot of other Google applications are also a part of our lives and we use them daily — Gmail, Android, Chrome, YouTube, Maps, Google Apps, etc.

These sets of products from Google each have more than 1 Billion users. Another similar example is Google Photos which sees around 1.2 billion photos and video uploads every year. Now imagine the kind and scale of infrastructure and services that are running at Google Data centers to power these applications to make sure that they support billions of users on a day-to-day basis.

 

Google Cloud

 

With Google Cloud Platform (GCP), Google is making the very same infrastructure & services that run these applications available to us and with these infrastructure and services, we should be able to run our applications like “Google” — seamlessly, efficiently and with massive scale.

But with other Cloud providers out there, the question arises — why should one choose Google Cloud over other clouds?

Let us dive into more about what makes Google Cloud different:

  1. Best-in-class Security: Google delivers end-to-end security solutions from the data center all the way to the device with a purpose-built infrastructure and security by default.
  2. Hybrid & Multi-Cloud: Google’s managed, cloud-native solution enables developers to write an application once, and then run it on-premises, on Google Cloud, or on other clouds with no change in infrastructure.
  3. Fully Managed No Ops: Google strives to make the cloud easy to use — automating manual tasks and configuration wherever possible so your teams can focus on development and less on managing infrastructure.
  4. Embedded AI & ML: Google embeds intelligence in everything, making it easy for everyone to apply artificial intelligence and machine learning in their own organization.
  5. Best of Google: They bring the best of Google in their solutions, and they can help to create a more innovative and collaborative culture in the organization.

Let’s come to 2 of the most important questions that drive the decision of which Cloud should I for –

  1. Does Google Cloud have all the required services to run my workload?
  2. How much does it cost?

Now even though Google came in a bit late to the market with cloud offerings than AWS & Azure, unlike them, Google didn’t have to do things from scratch. Google always had the underlying scalable, secured infrastructure and the services already in place and it was just a matter of exposing it to you.

My journey with the cloud started around 4 years back and I haven’t seen any other cloud grow as fast as how Google Cloud has grown over these years. Today they not only offer almost all the services that AWS & Azure are offering but also added certain services that Google was or still is the first or only one to offer like Cloud Spanner which is horizontally scalable RDBMS.

So, when it comes to the concern of whether Google Cloud will have the services to run my on-premise or AWS, or Azure workloads? The answer is yes.

Once this is clear, then the other question that remains is why would I choose Google Cloud over others. Even if Google Cloud is better, more secure, scalable and can support all my workloads, why should I move from my current cloud if things are already well over there?

Let’s see from the cost perspective — How do you save costs by migrating to Google Cloud?

The answer to this is quite simple — This is because of one — Google’s innovative and customer-centric pricing and two — providing better performance than other clouds in the same configuration setup.

Let’s see a few services as an example and then we will see a sample product architecture and compare its pricing against GCP, AWS & Azure. Let’s start with the most commonly used service — Compute Engine — This is IAAS offering from Google for running your VMs. This is similar to EC2 in AWS and Virtual Machines in Azure.

Compute Benefits

○ First to introduce per-minute billing and now billing by second.

○ Sustained Usage Discount: The more you use, the more you save. With SUD, you can save up to 30% without any prior commitment

○ Custom Machine Type: Unlike AWS & Azure where you can only choose the VM instances from a fixed catalog, GCP allows you to customize the amount of vCPUs and RAM as per your needs. This results in cost savings of up to 48%

○ Committed Usage Discounts: With CUD, you save up to 57% in VM costs too without any upfront payment.

○ Preemptible VMs: Migrate your offline and batch processes to preemptible VMs which are available at a much cheaper rate than On-Demand instances.

The comparison below will help you visualize the cost savings because of above mentioned features.

GCP vs AWS Instance Pricing for different workloads/instance-families

Block Storage Benefits

What type of disk (HDD/SSD) to opt for technically depends upon the IOPS (Input Output operations per second) and Throughput (Mbps) that your application needs.

Persistent Disks (PD) in GCP is a perfect example where Google Cloud is not just cheaper but also performance-wise quite better than other clouds.

The benefits of Block Storage in GCP are:

○ Cheaper Disk Price
○ Faster Disk Throughput
○ Higher Disk IOPS
○ Flexible Capacity Options

The cost differences between PD in GCP and EBS in AWS are:

PD vs EBS Pricing

As for the performance difference, please see the comparison of IOPS and Throughput between PD and EBS below:

 

Google Cloud disks offer higher IOPS and Throughput than AWS EBS for the same capacity. The difference is so high enough that we can even replace EBS SSD (1 TB) with higher capacity GCP PD (2 TB) to achieve the same performance at a much cheaper price.

Object Storage Benefits

○ Cheaper Pricing
○ Faster Data Retrieval Time

GCS vs S3 Pricing

These are just a few examples of certain services. A better way to compare the overall pricing will be to consider a production-level use case.

The below diagram is for a production-grade Mobile Back-end Architecture

Pricing as of June 2023

The pricing is calculated considering Mumbai region for all 3 clouds and is being referred from their respective pricing calculators. The links for the same as follows:
GCP: https://cloud.google.com/products/calculator/#id=15c2dad8-f1e1-4e58-944c-2d1bd300d2a1

AWS: https://calculator.s3.amazonaws.com/index.html#r=BOM&s=EC2&key=files/calc-464a6995797cd3d1d9e11896e77d83537b77904e&v=ver20200527oH

Azure: https://azure.com/e/7b049ca77e334324af6a6df37dc57d8a

You can see that the pricing for GCP is quite cheaper than AWS & Azure. Even at individual services/layers, GCP is mostly cheaper and for the service where it isn’t the cheapest, the difference is quite negligible.

Now, we have only covered mostly the infrastructure components for comparing the pricing and this is because they form a basis for most of the other services like Google Kubernetes  Engine (GKE) which is similar to EKS in AWS and AKS in Azure. So eventually the cost of all such similar services is cheaper in GCP than in AWS or Azure.

And this price difference is just out of pure lift and shift migration. You can further save more costs by performing a cost optimization process after the migration.

So what are you waiting for? Contact us and we will help you save and optimize your cloud costs on Google Cloud as well as assist you throughout the migration process.

Let’s start your journey with Google Cloud !!!