Comparing AWS and Azure

Two of the major players in cloud services today are AWS and Azure. They both offer software as a service or SaaS, platform as a service or PaaS, and infrastructure as a service or IaaS on a pay-as-you-go basis. But aside from being owned and operated by different companies, what really sets these cloud service providers apart? In this blog post, we will be diving deeper into the major differences in the services offered by AWS & Azure. But first, a brief overview of the two cloud giants.

Amazon Web Services is a cloud platform with over 175 services. It boasts more services than any other cloud provider. AWS services range from innovative technologies like analytics, compute, machine learning, augmented reality and block-chain, to databases, networking, security and content-delivery.

Initially released in early 2010 by Microsoft, Azure is significantly younger than AWS. Azure offers edge device services, serverless services, and infrastructure services much like its competitor, AWS. There are over 200 (featured at build) services listed on their website.

Points of Comparison

In order to contrast between Azure and AWS, we will be looking at these points:

1.) Ease of use

2.) Hybridization

3.) Compute (Virtual Machines)

4.) Databases

5.) Artificial Intelligence and Machine Learning Services

6.) Networking & Security

7.) Regions

Ease of use

Having used both cloud service providers, Azure seemed easier to get up and running which was not exactly the same experience I had with AWS. This appears to be the consensus across the board. However, due to its longevity and significantly larger user-base, AWS has more support and integration from third-party tools (many of which are open source) that greatly improve its ease of use.

Hybridization

Now, let’s consider an example of a large organization migrating to the cloud. This is typically a tough and tedious task which can be resolved with hybridization. Hybridization allows you to manage the cloud and your own physical resources such as data centers together as opposed to moving everything at once to the cloud. Azure appears to be the better of the two at hybrid clouds by making it easier to connect your data centers to the cloud. AWS, unsurprisingly, is hard at work to catch up to Azure in this aspect.

Compute (more specifically VMs)

Compute falls under the infrastructure as a service (IaaS) umbrella. It includes resources needed to run applications and computations such as CPUs, and RAM for virtual server hosting, application containerization, and serverless computing.

AWS offers two types of VMs: Elastic Compute Cloud (EC2) and Lightsail. Amazon Lightsail is designed for simpler applications and is also easy to use as it aims at introducing users to the AWS platform. On the other end of this fence, is EC2. Amazon offers a wide variety of general purpose EC2 types (known as “instance types” or simply “instances”) which are very flexible and can handle most use cases. In addition to this, Amazon also provides specially optimized instances such as compute optimized, memory optimized, accelerated computing, and storage optimized instances for efficiency

Azure, although not as flexible as AWS, seems to be winning the optimization battle. For instance, in addition to optimized instances similar to those of AWS, Azure adds a few extra such as instances optimized for in-memory hyper-threaded applications and high performance optimizations (I know right?). Azure also offers a special type of vm known as the DC-series which is designed to provide additional security of data by maintaining the confidentiality and integrity of data at rest, and during transmission. In my opinion, AWS seems to have a lot more flexibility but Azure has more optimized instances.

Databases

Both cloud service providers offer a variety of similar databases ranging from relational databases to NoSQL databases. Some of them come with cool features like Azure’s new SQL Edge which has built-in AI. Now, if you are used to a specific database like MariaDB, Cassandra or even MongoDB, AWS provides more popular database engine options and MongoDB compatibility with Amazon DocumentDB. However, due to the similarity in their offerings, you can still achieve the same tasks with either Azure or AWS databases.

Artificial Intelligence & Machine Learning Services

Both AWS and Azure offer ample AI and Machine Learning (ML) services.These fall under a fairly new concept known as Machine Learning as a service or MLaaS. On their websites, AWS lists 25 and Azure lists 37 of these services, ranging from generalized AI and ML to specialized ones like face and voice detection, speech translation, and so on. From the numbers alone Azure seems to be the better choice, however, that decision should be completely based on the application. I recommend you go through the list of AI and ML services offered by both cloud providers before making a decision about which one fits your needs.

Networking & Security

Networking and security is relatively similar in both platforms. The differences are merely syntactical. For example, in order to create a virtual network (VN) in AWS, a virtual private cloud (VPC) has to be provisioned first which is different from Azure where you can directly provision a VN. Azure also offers network services that are somewhat ready to use out of the box with little configuration. Services like the Azure bastion service which allows SSH and RDP connectivity to your virtual machine over TLS.

Regions / data centers

In cloud computing, a region is essentially a cluster of data centers. Based on the numbers from both cloud providers websites, Azure has more global reach. With over 60 regions (which is way more than double than that of AWS), Azure is available in more places globally than AWS. However, this doesn’t necessarily mean that Azure has more data centers than AWS.

In summary, the cloud services offered by Azure and AWS, and the features within them are so numerous that we merely scratched the surface with our points of comparison. The choice of which one to utilize will solely depend on the use case. In fact, the use case may not even matter. Thanks to competition in the cloud market, we will continue to see remarkable similarities in services offered by both providers. And migrating between the two platforms will eventually become hassle free.