Which Cloud is Best? AWS vs Azure
Before we get going with this one I ought to warn you it is a long one. So buckle up, and let’s get to it!
Microsoft Azure and Amazon Web Services (AWS) are the two biggest cloud providers according to Gartner. Moving to the cloud is becoming increasingly popular, and so Azure and AWS are continuing to grow. They’re each having to offer as much as possible to compete for new users.
So, which is better?
If you’re looking to find a cloud provider, for whatever reason, you’ve probably considered Azure and AWS. But there’s so much information about the two out there, it can be hard to choose which is right for you.
Let’s break it down.
We’ve split this comparison into sections so you can directly compare everything from the services offered, to the best price! Let’s get started with something simple.
File Storage Solutions - AWS vs Azure
One of the biggest benefits of the cloud is paperless storage solutions. It’s a great way to save physical space, time finding what you need, and prevent hassle sharing things with others.
Plenty of large businesses use cloud providers to store their data - and as you can imagine, they each have a lot that they need to store. So, when comparing Azure and AWS, this is definitely something to consider, especially if you’re looking to store a lot of important information on the cloud.
Let’s take a look at Azure first.
Azure provides a service aptly called Azure Storage. Within this is a number of useful resources, such as Azure Files, where you can replace or supplement on-premises file servers, transfer applications to the cloud, and simplify cloud development. Azure Files offers file shares in line with the industry standard SMB (Server Message Block) protocol, and can be made without managing hardware.
Because you can use Azure files to replace on-premises files, you can forget about power outage or network issues. Azure Storage also has a service to store text and binary data (Azure Blobs), a messaging store for messaging between application components (Azure Queues) and a NoSQL store for structured data (Azure Tables). On top of this Azure Storage includes scalable storage for data.
The data you upload to Azure Storage is accessible from anywhere globally over HTTP or HTTPS. There are different types of storage accounts available depending on your needs, each with different features and relevant pricing. So if you’re not going to use a huge amount of this storage capacity, you won’t have to pay the same as a much larger business that requires a lot more. You pay according to your needs. But more on this later.
Next we need to consider AWS.
There are a number of file storage solutions Amazon offers. It has a scalable simple storage service called Amazon S3. This can be used to store any amount of data, and has management features to help you organise your information.
Another storage service it offers is Amazon Elastic Block Store, which provides block storage volumes. Alongside this is the Amazon Elastic File System (EFS), an autoscaling service to provide your applications with the storage they need when demand increases. EFS stores data over multiple Availability Zones to increase availability and durability.
Another storage solution AWS provides is Amazon FSx for Windows File System, where you can move Windows-based applications that need file storage to your AWS cloud.
Amazon S3 Glacier is yet another storage solution they offer that provides low cost storage for data archiving, as low as $0.004 per gigabyte a month.
And the final storage solution it’s worth mentioning is AWS Storage Gateway, which lets your on-premise applications use AWS cloud storage. So, like Azure, AWS has a huge variety of file storage solutions.
It’s only fair to say that this is nothing more than a superficial look at both of these providers, and they both have such a vast array of different services it’s hard to pick a winner. But this wouldn’t be a proper comparison if we didn’t at least try. And so when it comes to file storage solutions the best cloud in our opinion is AWS for no reason more than the number of different services they provide. It’s close though
Data centers/Regions - AWS vs Azure
Let’s move on to datacenters and regions that each cloud provider covers. Although your data is accessible on the cloud, it still remains in one or more physical data centers.
Azure Datacentres and Regions
Azure gives you the option to replicate data across datacenters or geographical regions. This gives an extra layer of security against local issues or natural disasters.
They have 54 Azure regions, and Azure is available in 140 countries. Azure boasts more global regions than any other cloud provider, so are a good choice if you’re looking for maximum global coverage. This coverage can help you minimise the money and time spent maintaining a global infrastructure, whilst you keep a local presence.
These regions allow you to deploy apps wherever you choose. And they have plans to continue expanding these regions. Their website contains advice on choosing the best region for you, as each comes with various services, and can differ in pricing.
So how does AWS compare?
AWS has 64 availability zones that spread over 21 regions globally, but they have announced plans to create 12 more availability zones and 4 new regions. So they do have fewer than Azure, but they have plenty of plans to expand.
When choosing these new Availability Zones, a lot of thought goes into location to ensure risks of natural disasters and extreme weather are minimised. AWS availability zones are designed to protect performance and your data, even if their regions and data centers experience natural disasters, power outages, and more. Each region and availability zone is also isolated from the others to reinforce this resiliency. So data security is definitely a priority for AWS.
When it comes to Data Centres and Regions we’re going to give the win to Microsoft Azure.
Autoscaling Services - AWS vs Azure
Another key consideration when choosing your cloud provider is autoscaling services, scaling apps and hosting infrastructure to meet fluctuations in demand.
Azure tackles this with their service AutoScale, built into their cloud services, mobile services, virtual machines, and websites. Microsoft understands performance can be different in every application. So AutoScale is designed to scale apps that are CPU-bound, memory bound, or even scale it by a custom-made metric. Azure provides this scheduled autoscale to prevent traffic spikes from affecting your performance. You can schedule these auto-scaling targets yourself, based on traffic patterns, or other peaks you know are coming. Plus, Azure’s autoscaling can scale down your virtual machines when they aren’t in use, such as possibly at night, to save you money. You only need to use and pay for what you need.
As well as helping with all of this, Azure’s AutoScale monitors your application’s performance and can alert you when anything changes, or when AutoScale is triggered, so you remain in control.
AWS also provides an autoscaling service.
Amazon Elastic File System that we mentioned earlier is build to scale storage on demand without disrupting your applications. It is a fully managed service that uses a standard file system interface.
AWS autoscaling in general will recommend ways to optimise your performance and costs, to develop your business. You also have the opportunity to build scaling plans that automate how various resources respond to differences in demand. AWS autoscaling monitors your applications continuously to keep them operating at desired performance levels. When demand increases, this autoscaling will automatically increase resources to do this. Plus, like Azure, you only pay for what you need. AWS autoscaling in many applications is actually free, and it automatically removes excess resource capacity to save you even more money.
This is a very difficult one to call, but we have to give it to someone, and so we’re going for AWS simply because more people are familiar with how it works.
Ethics and Sustainability - AWS vs Azure
Ethics and sustainability might not be high on your list of priorities when choosing a cloud provider, but it’s actually a really key consideration for your long term use.
Azure achieved carbon neutrality in 2014. They also set and met a goal of averaging 1.125 power usage effectiveness for every new data center.
Microsoft itself is very committed to solving environmental issues. Some of the latest steps in their journey include a goal to cut operational carbon emissions by 75% by 2030. They’re increasingly turning to renewable energy as they expand Azure and their cloud infrastructure. Listing every step Microsoft are taking to ensure Azure and their other products are sustainable would take a long time. But they constantly update their progress online, for everyone to see!
So let’s move on to an overview of AWS’s sustainability. AWS is committed to achieving 100% renewable energy for their global infrastructure. In 2018, they exceeded 50% renewable energy, and therefore are well on their way to achieving this goal. They are continuing to build and use wind farms and solar farms across the world, with 3 new wind farms announced in April 2019.
AWS also highlight that moving from on-premise servers to the cloud causes a huge reduction in carbon emissions. On the cloud, server utilization rates increase, so businesses usually end up using much fewer servers than they would on-premises and therefore reduce the power required. No matter which cloud you choose, you’re almost guaranteed to become more sustainable.
Clearly both Azure and AWS have embraced sustainability as a core facet of their business, however there can only be one winner, and this time it’s Azure based on the wealth of projects Microsoft are funding around the world.
Machine Learning - AWS vs Azure
Another one of the major benefits of cloud providers is access to machine learning and AI services.
Azure Machine Learning
Azure’s solution is called the Machine Learning Studio, a browser-based server that requires no coding.
It’s a collaborative drag-and-drop tool that lets you build, test, and use predictive analytics models on your own data. If you’re new to machine learning, but eager to apply it to your business, you don’t need to worry. Microsoft provide various machine learning tutorials to help you get started.
Or, you can use their sample experiments. Azure’s machine learning is designed to improve your productivity and costs. So it’s definitely worth looking at if you’re moving to the cloud.
AWS Machine Learning
Developments of machine learning and AI on the cloud allow access to these technologies for every developer. AWS has a series of pre-made AI services that use computer vision, language, recommendations, and forecasting to improve productivity.
Alternatively you can use Amazon SageMaker to create, train, and use machine learning models, or build custom models for your data. When building your machine learning models, you can choose from a variety of algorithms, and once built can run your model on multiple configurations. AWS provide the same machine learning technology that they use themselves, allowing you to personalise experiences for customers, just as they do.
When it comes to Machine Learning the development work Microsoft have been doing, along with the familiarity of the interface and ease of integration means that they come out on top in this category.
Price - AWS vs Azure
Price is always going to be a key consideration when choosing a cloud provider. Even if you have plenty of money to dedicate to your project, you’ll want to get the best value. So let’s check out how Azure and AWS compare for pricing, starting with Azure.
Cost of Azure
Azure offer pay-as-you-go pricing to prevent you from spending more than you need to. If they aren’t cheaper, Azure also actually match AWS pricing for their most similar services. Plus, you can cancel at any time, meaning you aren’t held to long payment contracts.
Even within Azure, their services and tools analyse costs for you, to optimise your usage and prevent you from paying for unnecessary resources. Their website has a handy calculator, where you can predict how much you’ll be paying before actually spending any money. And as you only pay for what you use, you can take a look at the exact cost of the services you’re planning to use.
Many services, such as Machine Learning Studio that we looked at briefly earlier, come with tiered pricing. This usually includes a free tier, where you get slightly limited access, but the opportunity to try before you buy. Signing up for an Azure account currently gives you 12 months of free services, over 25 always-free services, and £150 of credit for other Azure services for a 30 day period.
Cost of AWS
AWS works on a similar basis. They operate a pay-as-you-go approach for their cloud services. You only pay for what you use, and there are no additional costs or termination fees when you no longer need the services. This helps you adapt to new business needs without losing unnecessary money.
There are some services that allow you to invest in reserved capacity, which can help you save up to 75% in comparison to on-demand capacity. Such as with Amazon EC2 and Amazon RDS. The more you spend here, the more you save. Volume-based discounts on AWS save you money the more you use. Like Azure, they have a calculator - the AWS Simple Monthly Calculator - to help you predict what you’ll be spending before you actually pay anything. They too have a free tier for many services, that lets you gain experience, trials, and more at no cost, for 12 months.
The pricing of any cloud provider can be quite complex. But both of these providers do a lot to help you minimise your expenditure. Sadly we have no choice but to call this one a draw.
Security - AWS vs Azure
Security of your data is always going to be important when choosing a cloud provider. There are so many measures that Azure and AWS take to keep your data safe that we won’t be able to cover everything in detail here. But we can start with a brief overview.
All of the data in Azure storage is encrypted. There are two types of encryption available: encryption at rest, and client-side encryption.
Azure gives you control over who can access your data and your security key. As we have seen, it also creates multiple copies of your data as a redundancy option in case any data centers are destroyed. They have a team of over 3,500 cybersecurity experts who are constantly working to safeguard your business and data.
Their cybersecurity intelligence will identify threats through machine learning, behavioural analytics, app-based intelligence, and the Microsoft Intelligent Security Graph. They have services to protect Distributed Denial of Service (DDoS) attacks, and the option to use Security Center to protect yourself against attacks. Their multi level security is designed to protect your data from any threat, including natural disasters.
AWS infrastructure is monitored 24/7, meaning your data is constantly protected.
Users are able to manage their data, with options to encrypt and move it if they choose. All traffic is automatically encrypted on AWS networks and between AWS facilities, to prevent any attacks. Like Azure, AWS replicates data and applications over multiple data centers in the same region by using their Availability Zones.
If you want to go even further than this, you can even replicate data over multiple regions. You always keep control over the region that holds your data. AWS even provides a service where you can report vulnerabilities, so you can take security into your own hands, past their support network of guidance and advisors.
This is another very difficult category to call, and so I’m not going to. It’s another draw.
Customers - AWS vs Azure
Finally, let’s compare the customers that are currently using Azure and AWS. This won’t necessarily impact your choice. But, it’s a good way of seeing the sorts of companies that work well with the systems you are potentially going to be using. Especially if you’re worried about how your industry will fare on the cloud.
Some of the well-known companies that use Azure include: Johnson Controls, Fabletics, Seattle Seahawks, Adobe, and HP, among others. The list is pretty huge, so you’re bound to find examples from the same industry as you.
AWS also has some very familiar companies on their cloud service. Some of the biggest names includes: 21st Century Fox, Airbnb, AOL, BMW, and NASA. Can you see yourself adding your name to this list?
They are more well established and have some seriously successful customers utilising both services, but AWS takes this one. Azure isn’t far behind, and this may change in the future, but right now the winner is AWS.
Now I have the somewhat tricky task of trying to summarise this comparison. It’s instantly clear that both AWS and Microsoft Azure provide a vast array of services, and there is feature parity at pretty much every stage.
There are things one does better than the other, and there are things that are easier on one cloud than the other. But the fact remains that it it genuinely impossible to state with any accuracy which is the better cloud between AWS and Azure.
What we can do however, is say which of the two is best for you and your specific business requirements. So if you’re still struggling to decide whether you should be looking at Azure or AWS why not give us a shout and we can help you decide.