Determining the right cloud platform for your business

Cloud has officially entered the commodity age; yet there are many disparities in how hyperscalers identify their services, although essentially, they are doing the same thing. They offer analytics, AI, IoT, Serverless, K8s, Hybrid, IaaS, PaaS, and other services, but their implementations and delivery methodologies may differ greatly.

Let’s examine this with a few examples:

• AWS is maturing on application level-specific functions such big data, security, and data processing, evolving and rising bottom-up from simply hosting any IT infrastructure.
• Google was one of the first companies to integrate data analytics and related platforms with desktop/user interface services.
• Microsoft merged all market-focused technologies to transform business applications, including system functionality such as desktops and mobiles.

As industry leaders follow closely on the heels of each other with a subtle war on pricing, it is difficult to zero in on one single platform that is best suited to your business. When selecting a public cloud provider, a thorough evaluation is necessary to determine how mature the services are and if they are part of your company's long-term growth plans.

The big-tech companies do have a comprehensive cloud portfolio for quick savings and digitalization. With unique features and rapidly emerging technology, selecting the best supplier or being able to simply move to another would be more vital. The current challenge for most organizations is to find the sweet spot for long-term growth and a sustainable cost-benefit ratio between features and functions.

The growth of cloud platforms

According to Gartner's Magic Quadrant report on public cloud providers, the capability gap between hyperscale cloud providers has begun to narrow; however, fierce competition for enterprise workloads extends to secondary markets worldwide. Indeed, the financials from AWS, Microsoft Azure, and Google Cloud have all been strong.

AWS has been solidifying its place as a strong contender for edge use and hybrid cloud cases. In both database PaaS and IaaS products, it has the largest share of the global market. Microsoft is next in the list of leaders, but it’s not far behind. Microsoft Azure offers a comprehensive collection of end-to-end solutions for a wide range of applications and workloads, making it ideal for Microsoft-centric businesses. In the application developer PaaS segment, the tech giant leads the hyperscale cloud providers, concentrating on open-source applications with tools like GitHub and Azure DevOps. Completing the group, Google’s open-source contributions like TensorFlow and Kubernetes have transformed enterprise IT. Through its Google Cloud Platform, Google has seen a significant rise in its market share of DB, PaaS, and IaaS, due to improved alliances with telco companies and the expansion of its hybrid capabilities (GCP).

While we look at the top public cloud service providers' revenue and market share, spending continued to consolidate in 2020, propelled forward by an ever-expanding SaaS and PaaS portfolio.

• According to Statista, global cloud infrastructure service revenues totaled $39 billion in Q1 2021, taking the year-to-date figure to $150 billion: Here
• Amazon’s market share in the worldwide cloud infrastructure market amounted to 32% in the fourth quarter of 2020, still exceeding the combined market share of its two largest competitors, Microsoft (20%), and Google (9%): here.

Enterprises are increasingly seeing the cloud as a digital transformation engine as well as a technology that improves business continuity. While enterprises are deploying more multi-cloud arrangements, the digital budgets are increasingly going to cloud giants.

A closer look at competitive pricing models

Each cloud service provider has its own pricing model, aiming to provide good services to customers at a reasonable cost. Each cloud platform has the potential to influence pricing, even before considering how overall pricing will be determined, whether it is based on time, volume, priority pricing, content, location, server type, or any other metric the cloud provider wishes to promote.

Here are some of the most common pricing models to consider when choosing a provider or service.

Rewarding loyalty in committed usage

There are several factors to consider when it comes to cloud migration costs, including region, vCPU, memory, OS, on-Demand, and discounted instances. With a one-year commitment and no upfront cost, the following are the cost considerations for AWS, Azure, and GCP. The cost of Google's memory optimized instances is the highest, but the cost of a 1-year commitment is the lowest.

Table 1: Cloud Pricing Comparison 2021: AWS vs Azure vs Google Cloud (simform.com)

Time is literally money with per second billing

To attract more customers and provide more price flexibility on compute usage, cloud service providers have switched to a per second billing. This has benefited consumers by allowing them to use instances for a shorter period.

AWS – In 2017, AWS announced that each partial instance consumed in an hour would be invoiced per second for all Linux instances.

Azure – In 2019, they began billing by the second, but they are still listed as being billed by the minute.

GCP – Per second billing on GCP is preferred over AWS because it applies to all VM-based instances, whereas AWS only does so for Linux instances.

Finally, the per second billing model is significant in that pricing flexibility is a key feature of cloud computing: the bill is much closer to actual usage, making costs more predictable.

Serving it serverless

Serverless computing is a cloud architecture that allows you to delegate more operational responsibilities to cloud providers, increasing your agility and innovation. AWS Lambda, Azure Functions, and Google Cloud Functions are some of the serverless services available. The processing power you use is billed in 100-millisecond increments by these cloud services. Thus, serverless allows you to create and run apps and services without having to worry about servers. Given their free tier services, AWS, Azure and GCP’s serverless pricing is nearly identical.

In conclusion, as pricing models change, businesses must be aware of transaction-based cost models. Data in a rest mode has progressed to the point where price elements are now comparable. The associated pricing for data in transit (transactions, ingress, and egress) are the hidden cash cows, as most are volume-related, with exponential potential growth. This cost factor currently makes up a small portion of total housing costs, but it has the potential to skyrocket in the near future.

Stay tuned to this space to learn more about how cloud can transform your business capabilities and help you along your digital transformation journey.

Read my previous blogs in this series:

Multi-cloud trends and strategy: An executive summary - Atos

Decoding the Cloud Computing timeline

Topics

• Cloud
• Tech Insights