Choosing a cloud provider for your business can be extremely difficult and important decision.
It hugely affect how well and smoothly your website and applications run. As well as how much you’re expected to pay for the luxury of using it.
Of course, there isn’t an all-fit-all solution all there.
However, there are four big players on the market – AWS, Azure, Google Cloud and Rackspace.
In this blog post, we’re going to make a detailed comparison between all of them based on factors such as –
pricing, features, implementation, compute, storage, databases and networking, downtime and customers.
Pricing
AWS, Azure, Google Cloud and Rackspace all have competitive pricing strategies.
They’re all putting great effort into taking workloads and migrating them to the cloud.
In the past, you would be able to only pay based on your monthly usage, but now you can also pay per minute or even per second.
Not only that, but there are great discounting options that happen with committing to usage on the cloud provider for a one-year or three-year period.
Let’s look at the differences in pricing between all cloud providers:
AWS – the pricing level of AWS is easily presented from SMB to Enterprise. One of its biggest benefits is that you don’t have to buy a quickly outdated private cloud solution.
You just need to note your activities, and the cloud will adjust accordingly.
If you just want to put some data in the cloud, their three-tier model for storage is very helpful. The only bad thing is that the percentage price difference between 50TB to 500TB is not that big.
Here’s some helpful resources you should check out about pricing:
It’s important to notice that the levels of cloud tiers do vary greatly and it’s good to compare them against your requirement.
Azure – there are a lot of compelling elements to Azure when it comes to pricing.
Their pricing strategy is more competitive in certain section than Amazon and Google because of their goal to lead segments of the cloud.
Here’s some helpful resources you should check out about pricing:
The TCO calculator will help you understand how to lower the total cost of ownership of your on-premises infrastructure. At the same time, you would be able to calculate the cost savings of migrating application workloads to Microsoft Azure.
Google Cloud – its core goals is to have pricing that competes with its core competitors while charging you for your exact usage.
Among the core benefits are no upfront costs, pay-as-you-go and no termination fees.
Here’s some helpful resources you should check out about pricing:
Rackspace – it comes evenly priced with other cloud solutions even though it’s slightly more expensive than AWS.
With Rackspace you’re also a bit limited when it comes to the choice of bundled services with fixed prices.
When it comes to storage, Rackspace offers a better deal for SME’s. They have really good deals for middle-of-the-road servers.
Even better deals you can find for the low-end and very high end of the markets. This means for individuals and tiny startups or for large corporations.
The good thing about Rackspace is that they only charge you for the amount of storage you actually use, and no charge for I/O activity. Unlike AWS.
Features
Choosing a cloud provider to a huge extend comes down to the needs and wants of each individual customer.
In general, ASW, Azure, Google Cloud and Rackspace offer largely similar basic capabilities around flexible compute, storage and networking.
They share common elements of the public cloud – self-service and instant provisioning, auto scaling, plus security, compliance and identity management features.
All of them invest and innovate heavily in their cloud services and are backed up by big companies that allow them to do that.
AWS features – AWS offers the largest set of services with nearly 100 across compute, storage, database, analytics, networking, mobile, developer tools, management tools, IoT, security and enterprise applications.
This is mainly due to the fact they are the oldest cloud service provider among the group.
Back in April 2015, Amazon launched the Amazon Machine Learning service to help developers create machine learning models.
Later in 2016, they announced three new machine learning services for image recognition (AWS Rekognition), text to speech deep learning models (Polly) and the engine that powers Alexa (Lex).
Azure features – Microsoft provides a great set of features for its Azure users. The good thing is they also provide more capabilities based on the number of users.
Their Enterprise Agreements [EA] enables large organizations to earn benefits from committing usage to Azure.
When it comes to their degree of certification they claim that they have more than any other cloud provider.
Azure also has a Machine Learning Studio that allows developers to write, test and deploy algorithms, as well as a marketplace for off-the-shelf APIs.
Google Cloud features – Google Cloud has a Cloud Machine Learning Engine, created to help machine learning engineers build models based on its open source TensorFlow deep learning library.
They also have extremely helpful off-the-shelf APIs for things like natural language processing, translation and computer vision.
Google Cloud is really big when it comes to partnerships, allowing customers to run various apps and services in their cloud environments.
For example: they have a key partnership with vendors like SAP, Pivotal and Rackspace.
Rackspace features – Rackspace offers 17 types of Linux-based instances and 16 Windows-based instances.
They’re organized in the following categories:
- General Purpose,
- Compute Optimized,
- I/O Optimized
- Memory Optimized.
You also have the option to use Windows instance types with built-in SQL servers. Last but not least, you can also use Vyatta Router instances that are really good for setting up network infrastructure.
The bad thing about Rackspace is that you would need to get help from a third party for all the features.
Rackspace also doesn’t natively support autoscaling. In order to make it work, you should turn to RightScale.
RightScale allows you to manage many types of servers on any cloud provider given that you load up their “agent” on your servers.
Implementation
Besides just features and pricing, it’s important to also compare cloud providers based on how easy it is to get started with them.
Now let’s take a look:
AWS implementation – AWS has a great getting started page that makes it super easy to start using their services.
The good thing is that they break it down by platform.
The following platforms are available:
- Android
- iOS
- Browsers
- Java
- .Net
- Node.js
- PHP
- Python
- Ruby
In all cases, when you get started they provide you with a sample code so you can start the integration.
Azure implementation – starting with Azure is extremely easy as well.
The good thing about the process that they follow is that they think both for developers and operations, and have some great in-depth guides for that.
What’s really cool is that they are an easy forward to the relevant department if you need to check suitability.
You can check out the guides here:
You can check out the guides here:
Google Cloud implementation – they allow you to easily get started with their Cloud Launcher.
Good thing is they provide you with some really good documentation.
They offer ready-to-go development stacks, solutions, and services to accelerate development.
You would be able to choose among:
- Popular solutions (WordPress, LAMP Stack, MongoDB, Cassandra, Redis)
- Database (Redis, Cassandra, DataStax Enterprise, EDB Postgres Enterprise)
- Blogs and CRM (WordPress, Drupal, Joomla, TYPO3)
- Development tools (Redmine, Jenkins, GitLab)
Rackspace implementation – You can use the Cloud Control Panel to manage your Rackspace Cloud infrastructure through a simple and intuitive interface.
Even though it’s not as easily to use as other platforms, it does provide a lot of options.
On the top of the panel you can see your product-area headings (Servers, Orchestration, Networking, Storage, Databases, and Backups).
The products on the other hand are categories as follows:
- Servers
- Orchestration
- Networking
- Storage
- Databases
- Backup
There are also some cool features that help you organize, find, and efficiently manage your resources:
- Tags (customized labels that you can add to any of the resources in your cloud infrastructure)
- Filters (with them you can find all resources that share the following common properties: tag, status, image, RAM, region, or type)
- Search (with its help you can locate assets by using asset name, tag, or IP address)
With the control panel you can also perform a variety of administrative tasks. For that you can use the following:
- Gear icon (you can see it to the left of a resource name which provides a drop-down menu with all available actions for that resources)
- Actions menu (upper right corner of a page when looking at details for a resource)
- Edit pencils and options (you can use the pencils or option text to edit the existing setting for many resources)
- Pop-up boxes (you can use it to modify the settings or perform an action without going to another page)
Compute, storage, databases and networking
Another important comparison we should make between AWS, Azure, Google Cloud and Rackspace is when it comes to compute, storage, databases and networking.
Let’s see the differences:
AWS – when it comes to compute, Amazon’s main offering is its EC2 instances that comes with a large number of tailored options.
They also have related services such as Elastic Beanstalk for app deployment, the EC2 Container service, AWS Lambda and Auto scaling.
When it comes to storage, Amazon includes its Simple Storage (S3), Elastic Block Storage (EBS), Elastic File System (EFS), Import/Export large volume data transfer service, Glacier archive backup and Storage Gateway, which integrates with on premise environments.
Azure – its main computing offer comes down to its Virtual Machines.
They come with extra tools such as Cloud Services and Resource Manager to help deploy applications on the cloud. As well as their Azure Auto scaling service.
When it comes to storage, Azure offers its core Azure Storage service, Azure Blob block storage, as well as Table, Queue and File storage.
Additionally, Microsoft also offers you Site Recovery, Import Export and Azure Backup.
Google Cloud – when it comes to computing, Google offers scalable Compute Engine that delivers VMs in Google’s data centers.
They are extremely quick to boot, come with persistent disk storage, promise consistent performance and are highly customizable depending on the needs of the customer.
Rackspace – their cloud servers are compatible to EC2. It is the most customizable platform in the list and the closest to running a private server.
The development tools are great and if you want to work in a familiar environment with as much customizability as you want this is the right server for you.
Downtime
When customers choose a cloud provider they usually expect to be able to access their files all the time from anywhere around the world.
However, even big cloud providers occasionally go offline for short periods so they can maintain their systems.
That’s why it’s also good to compare AWS, Azure, Google Cloud and Rackspace when it comes to their downtime.
AWS – Back in 2014, Amazon had to reboot around 10% of their cloud services. This was caused by a major vulnerability in the open source platform — Xen hypervisor.
In this case, AWS was down only for the duration of the reboot so they can install patched on the system.
While there were several areas affected by the downtime, the total were 23 outages for a total of 2.69 hours.
Azure – back in 2014 Azure had a hard hit. Besides their usual maintenance, they had some serious disruption in November.
As a result customers experienced intermittent connectivity issues. This seriously impacted the storage platform.
During that time there were 92 outages on the compute sphere. This lead to customers not being able to access their virtual machines or perform any work for combined 39.77 hours.
The storage service was affected as well with 141 outages for a total of 10.97 hours.
In total for the two platforms, Azure suffered 50.74 hours in which users were not able to retrieve their data.
Google Cloud – they also had their downtime period. Unlike other companies, Google didn’t give specific reasons for their downtime in 2014.
The reason might be that the downtime wasn’t caused by a single one problem.
For example back in 2014, there were several big online vulnerabilities that caused problems for cloud companies.
This caused the need for patches and reboots so they can be fixed.
The good thing is that the interruptions were minimal and their cloud was down for only combined 4.46 hours throughout the year.
Rackspace – Rackspace also had some rebooting issues in the past. They had the same vulnerability with the Xen hypervisor open source program like AWS.
However, they needed to restart more servers to fix the problem. Around 10% of their servers were rebooted to install the patch.
The good thing is Rackspace got creative. They staggered the downtime in their coverage area so that no two regions would be down at the same time.
That way they only had 7.52 hours of down time while they fixed the problem.
Customers
The type of customers that use a certain cloud provider might not be the main selling point. However, they can be a key indicator that can show you what’s possible with its service.
AWS – it has a huge portfolio of high-profile clients. Back in 2013, CIA chose AWS to build its private cloud.
Another huge customer of AWS is Netflix. Back in 2016, they decided to shut down their own data centers and move to the cloud.
AWS has been successful when it comes to convincing a lot of traditional businesses to move to the cloud.
This includes AstraZeneca, NewsCorp, AirBnB, Aon, Channel 4, Financial Times, Dow Jones, Kurt Geiger, Lonely Planet, Nasdaq, Nike, Nisa Retail, Pfizer, and the Royal Opera House.
Azure – it has less high-profile clients than AWS and it is mainly focused in the SaaS space.
They also have some more well-known customers such as Pearson, Ford, NBC News and Easyjet.
Google Cloud – they have some high-profile clients such as UK bank HSBC which chose them for their analytics and machine learning capabilities.
Snapchat and its parent company Snap that uses Google Cloud for IaaS, but they also use AWS.
Other big clients are Home Depot and Disney.
Rackspace – GitHub is among the most famous Rackspace customers that moved to Rackspace back in 2009.
They have well-representative brands in industries such as
- aerospace & defense,
- appeal, automotive,
- business services,
- construction,
- customer goods & services,
- education,
- energy & utilities,
- financial services,
- food & beverages,
- furniture, hardware,
- healthcare & medical,
- hospitality & travel,
- manufacturing,
- media & entertainment,
- non-profit,
- pharmaceuticals,
- printing and publishing.
Other well-known brands are Basic Outfiters, Mazda, Sesame Street, Coinstar, Domino’s, McKesson, Fujitsu, Omaze, Pipedrive, etc.
Pros and Cons
Last but not least, let’s summarize the pros and cons of each cloud provider:
AWS – a great benefit of AWS is that they started building its suite of cloud services since 2006. They are all enterprise-friendly so they’re liked very much by CIOs and developers.
Among AWS strongest points are its platform configuration options, monitoring and policy features, security and reliability.
Their general product strategy tend to dominate the market and they also have an extremely well-developed partner ecosystem. This allows it to build their AWS marketplace with a huge number of third-party software services.
Really big advantages of AWS are also its openness and flexibility.
One of the cons of AWS is its hybrid cloud strategy. AWS has tended to be dismissive of the benefits of on premise private clouds.
The problem is that many organizations prefer to keep sensitive data within their own data centers. And then use public cloud for other purposes.
Another con of AWS is the scale of its offering. It could be hard at times to navigate the large numbers of features that are on offer. This make some people believe AWS is a hard cloud server to manage.
Azure – the big advantage here happens when Microsoft already has a strong influence within an organization. This is when they can easily help these companies easily transition to the cloud.
Azure easily connects with other key Microsoft on premise systems such as Windows server, System Center and Active Directory.
A key strength for Azure is its PaaS capabilities, even though AWS also has them.
A serious con is that Azure had a series of outages over the years. That’s why a lot of experts recommend considering disaster recovery capabilities away from Azure for critical applications hosted in the cloud.
Another bad thing about Azure is that it is very limiting when it comes to options to support other platforms.
So it will be extremely hard to run anything else than Windows server. Even though Microsoft has put some effort into accepting open source platforms.
Google Cloud – it has a good track record working with innovative cloud-native companies and with the open source community. However, they struggled to get into the enterprise market.
They have mostly focused on proving itself on smaller, innovative projects at large organizations, rather than becoming a strategic cloud partner.
Google would need to work on their partnerships and support pre-cloud businesses so they can attract enterprise level clients.
A huge advantage of Google Cloud is its machine learning tools, with the company’s internal AI expertise and popular TensorFlow framework.
Rackspace – one of the major cons for Rackspace is its flexible pricing and plans. And that combined with some good features and functionality.
They also have an obsessive focus on customer service. Rackspace has the ability to offload your cloud storage system management so you don’t need to hire a sys admin.
One of the things that suck is that there are no interoperability tools between using or switching from one vendor or another.
The other thing is that it is less convenient and speedy than having an in-house sys admin team.
Which is your proffered cloud provider and why? Let us know in the comments below.