A High-level Comparison of AWS, GCP, and Azure Services

Are you new to the cloud and want to learn more? As long as you have some technical background, this article will help give you an introduction to what the Cloud is and the major cloud service providers in the current market. Some of the most popular services each cloud provider offers will be described on a high level throughout the article. Subsequent articles will go into much more detailed comparisons within each core element of the cloud.

What is the Cloud?

Cloud Computing is one of the biggest trends in the Tech Industry today. What is Cloud Computing? According to Cloudflare, “The cloud” refers to servers that are accessed over the internet, and the software and databases that run on those servers. One of the reasons for the cloud’s popularity is because it allows customers to purchase services in a pay-as-you-go model. Customers don’t have to buy and maintain their own infrastructure which saves upfront capital expenditures and speed of deployment. The Cloud is versatile, as it offers high-speed access, high availability, and elasticity. Elasticity refers to the resources in the cloud that can increase and decrease based on the amount of information needed to process.

Who are the Major Providers in the Cloud?

Cloud is an appreciable business model, but who are the top players in the game right now? The 3 major cloud providers currently are AWS (Amazon Web Services) which is Amazon’s cloud computing platform, Azure which is Microsoft’s cloud computing platform, and GCP which is Google’s cloud computing platform.

AWS leads the way as it offers the most diverse suite of services, tools, and support while having the largest market share. According to Canalys, AWS holds 32% of the market share, far ahead of the competition. Behind it, is Azure which currently holds 19% of the market share. Even though there is a sizable gap between AWS and Azure, Azure has gained some speed and closed the gap from previously. A far third is GCP (Google Cloud Platform) which holds just 7% of the market. GCP may be at the bottom of the trio but it still provides some fantastic services.

There are several reasons for AWS’ lead in the market. One of the main reasons is that it was one of the earliest Cloud Computing Companies, Amazon launched AWS in 2006 and was one of the main driving forces in popularizing the cloud. AWS also boasts a significant amount of services, AWS currently offers more than 175 featured services with a diverse array of functionality. AWS’ global footprint is quite impressive as they currently host 24 regions around the globe along with 77 availability zones. Regions are actual, real-life geographic location where your public cloud resources are located. Availability Zones are isolated locations within regions from which public cloud services originate and operate. They plan to increase those regions to 30 and add 18 more availability zones!

Next up is Azure, which comes in at a strong second. Microsoft released Azure in 2010 originally as Windows Azure, later being renamed to Microsoft Azure. Although they were a bit behind AWS in jumping on the cloud train, Azure has come a long way and is a strong contender as a cloud provider. Azure also boasts a large amount of services as it’s cloud platform offers more than 200 products and services. Azure also has the biggest global footprint out of the three major cloud providers with over 60 plus regions.

Last but not least is GCP, originally launched by Google in 2008 with one of it’s more famous services called App Engine. Since then, GCP has come a long way and now gives customers a chance to explore over 90 products. It is definitely far behind AWS and Azure in the amount of services it offers for the cloud and the functionality amongst them, but some of the services GCP offers are very useful. GCP has 24 regions around the globe with 73 zones; which is comparable to AWS.

What are the Different Compute Services in AWS, Azure, and GCP?

Let’s move on to talking about some of the popular core elements in the cloud and what each cloud provider offers. Some of the biggest domains in the cloud are Compute, Storage/Databases, Networking, and Security. We shall delve into Compute first.

Depending on the needs, you are able to choose from services that give you more control over services that are fully managed by the cloud provider. One of the hottest trends today is a term called serverless. This is where users do not have access to the underlying infrastructure because the Cloud Provider spins up and down the infrastructure based on demand, completely removing management from the user’s end. This technology saves companies a lot of time and money. Another popular trend in the tech industry right now is the use of containers and container orchestration. Containers are a method to package an application so it can be run, with its dependencies, isolated from other processes. Container orchestration is open-source technology known as Kubernetes which allows users to orchestrate their container workloads. It is heavily used and becoming increasingly popular.

AWS

AWS is difficult to beat as it has quite a range of compute services. Some of the most used popular services include EC2, Elastic Beanstalk, and Lambda. EC2 (Elastic Compute Cloud) is a critical componentcore of the Cloud. It is AWS’ IaaS (Infrastructure as a Service) and allows the most flexibility and control to customers. You can manually provision and scale as much computing resources ______ as you need as well as select your underlying operating system. Elastic Beanstalk is AWS’ PaaS (Platform as a Service). This allows customers to deploy web applications and developers to focus on writing code while not having to worry about provisioning the infrastructure resources. By configuring some basic parameters, Elastic Beanstalk will automatically provision the resources and handle the operations (such as load balancing) while giving control of the underlying infrastructure (such as the ec2 instances that you can access at any time). Lambda is AWS’ serverless compute offering which allows users to simply focus on writing the function code. AWS offers quite an array of Container Orchestration Services such as ECS(Elastic Container Service) which is AWS own container orchestration service but they also support Kubernetes with EKS(Elastic Kubernetes Service). With EKS, you can get the advantage of Kubernetes while decreasing the management time of the underlying infrastructure. If you really want to go all out and go serverless, AWS also has the option of Fargate to deploy containers and the resources will automatically scale up and down based on demand.

Azure

Azure also provides a solid array of compute options, most of them similar to the ones AWS has. Azure has Virtual Machines, App Service, and Azure Functions. Virtual Machines are Azure’s offering of IaaS and they allow the customer the most flexibility. App Service is Azure’s PaaS Compute offering and allows users to easily deploy their web and mobile applications without worrying about provisioning the infrastructure. Azure functions are Azure’s serverless compute offering and will also scale up and down based on the workload. In terms of Container Orchestration, Azure has some well-defined services. They offer Container Instances which is one of the easiest ways to run your application in a container on the cloud and it is also serverless. They also have AKS (Azure Kubernetes Service) which is Azure’s managed Kubernetes service and allows users to use the features of Kubernetes on the cloud.

GCP

GCP Compute IaaS offering is called Compute Engine. Compute Engine comes with a ton of features including some notable ones dedicated to Machine Learning. GCP PaaS offering is called App Engine and is offered as a managed service and unmanaged. App Engine ‘Standard’ is completely serverless and scales on-demand with no infrastructure to manage. If you would like some more control of your Infrastructure, App Engine ‘Flexible’ is the way to go as it is most similar to AWS Elastic Beanstalk. GCP also has a serverless function offering with a service called Cloud Functions. GCP has some great Container Orchestration Resources with GKE (Google Kubernetes Engine) and Cloud Run. Google originally designed Kubernetes and GKE is one of the best managed Kubernetes systems out there. Cloud Run is a serverless offering that allows you to deploy container images and have the infrastructure scale on-demand to the workload. Cloud Run consists of two deployment platforms, Cloud Run, which is used without Kubernetes and Cloud Run for Anthos, which runs on GKE.

What are the Different Database/Storage Services in AWS, Azure, and GCP?

AWS

AWS offers a variety of database/storage services to choose from. Their most popular storage service is S3 which has a virtually unlimited capacity. S3 can store images, large videos, and host websites too! It comes with numerous security features from fine-grained access control to different levels of encryption. It has various tiers that allow customers to save on money while selecting the best tier for them like archiving their data with Glacier. AWS also has EFS (Elastic File System) for high-speed access file storage and storage gateway for hybrid cloud storage. For Databases, AWS has its main relational database service RDS which supports many database systems such as MySQL, PostgreSQL, MariaDB. RDS has a cloud-optimized version called Aurora which gives much higher performance to the customer. DynamoDB is their fully managed NoSQL database with high-speed performance and an abundance of features that…. allow customers to use DynamoDB as their key-value and document database for mobile, web, gaming, ad tech, IoT, and other applications that need low-latency data access at any scale.

Azure

Azure brings some significant storage services. Their storage services are all combined into a platform called Azure Storage. It consists of Azure Blobs, Tables, Files, Queues, and Disks. Azure structured their storage in a different way from AWS but you are able to store a ton of data and scale as much as you like. The customer has the option of (selecting the amount of availability) being as highly available that they deem necessary. Azure storage provides Locally Redundant Storage has the least availability as it only replicates within a single physical location to Geo-Zone-Redundant Storage. Microsoft recommends using GZRS for applications requiring maximum consistency, durability, and availability, excellent performance, and resilience for disaster recovery. Azure Blobs has an archive feature similar to that of S3 Glacier from AWS. Azure File storage can be used to store and share files. Azure File Storage allows applications to mount file shares from anywhere in the world, your on-premises applications can take advantage of cloud storage without change. Azure’s database services are structured differently from AWS. Azure has different managed services for different database systems such as Azure Database for PostgreSQL and Azure Database for MariaDB whereas AWS combines all of them into RDS. Azure has an amazing managed SQL server platform as (because) SQL server is a native Microsoft product which is designed to deliver performance, security, and innovation. Azure has it’s high-performance NoSQL database types combined into a service called Azure Cosmos DB. Azure Cosmos DB is a fully managed NoSQL database service for modern app development. Get guaranteed single-digit millisecond response times and 99.999-percent availability, and open-source APIs for MongoDB and Cassandra.

GCP

GCP has 3 main storage services including Google Cloud Storage which is very similar to AWS S3 as it has unlimited capacity with very high durability. It comes with an abundance of features including the ability to archive data and Multiple Redundancy options where you can store data across multiple locations. GCP has Google Cloud Filestore which is great for storing and accessing files. GCP comes with some awesome database options. Google Cloud SQL is a great option to host MySQL, PostgreSQL, or SQL server managed databases and if you want to scale even more and achieve high availability with fast access across the globe, GCP has a great option with Cloud Spanner. Although, GCP does lack in supporting the amount of database management systems that AWS and Azure provide. GCP has its NoSQL database service known as Bigtable which is great for high-performance workloads similar to AWS DynamoDB.

What are the Different Networking Services in AWS, Azure, and GCP?

AWS

AWS has the most core networking services such as Elastic Load Balancing in which they have multiple types of load balancers fit for a customers’ needs. A user could use an Application Load balancer for their application hosted on AWS or a network load balancer for high-speed TCP connections. AWS has its own VPC(Virtual Private Cloud) with various managed tools that can be used such as NAT gateways which allows your private subnet to connect to the internet securely. It hosts Transit Gateway which is amazing for the complex networks that large enterprises work with. AWS can also send information throughout the globe at high speeds with Cloudfront and uses Route53 for DNS workloads.

Azure

Azure contributes a great deal in the networking and content delivery realm as it provides networking and content delivery services to meet customer needs. Azure has load balancing services, which vary from AWS. Azure has it’s Azure Load Balancer which works at layer 4 on the networking layer and then the Application Gateway which works at layer 7 of the networking layer. Azure has a Virtual Network that can host many cloud resources privately and securely. Many tools are integrated with the Virtual Network as well. Azure has a content delivery network called Azure CDN which delivers content at a high speed to users all over the world and Azure DNS for DNS workloads.

GCP

GCP contains various networking services fairly like Azure and AWS. You can load balance with GCPs Cloud Load Balancing Service. GCP definitely gains some points here as you can load balance across the globe as well as internally. You can configure a VPC network to host all your cloud services privately and use tools such as Cloud NAT for the resources in the private subnet to gain access to the internet privately. Cloud CDN provides high-speed content to users around the globe and Cloud DNS for DNS workloads.

What are the Different Security Services in AWS, Azure, and GCP?

AWS

Security is one of the most important aspects of the Cloud. AWS does a great job with this as they have some useful security services that can prevent disasters from occurring. AWS has some very popular services such as IAM which is central for managing permissions and access in the cloud. It also has a service to protect your applications from attacks with AWS Web Application Firewall and your applications from DDOS Attacks with AWS Shield. Shield has a standard and advanced tier. The advanced tier is significantly more expensive at $3,000 a month but provides a lot more protection than the standard tier. AWS also comes with KMS in order to manage encryption keys.

Azure

Active Directory is one of the most popular systems used for access management and Azure has a highly integrated Cloud Active Directory Service as Microsoft owns the Active Directory product. Azure also comes with a multitude of security services such as Azure DDOS protection which protects applications from DDOS attacks and Web Application Firewall(WAF) which protects applications from web attacks. Azure’s Key Vault is a great way to store and manage keys.

GCP

GCP consists of standard security services like IAM for controlling access and permission within the cloud. GCP has a super intriguing service called Cloud Armor which has both DDOS protection as well as WAF for applications. GCP’s Cloud Key Management provides the ability to store and manage encryption keys securely.

Closing Remarks

In conclusion, the three major cloud providers all provide valuable services in each core cloud element. AWS is currently ahead in the market in terms of market share and GCP is a little behind the other two cloud providers. This was a high-level overview of what each cloud provider has in terms of services and capabilities at the time but Briebug will be releasing subsequent articles that will be going more in-depth comparing each cloud computing element (Compute, Databases/Storage, Networking, Security) and each of the services, capabilities, and pricing that the cloud providers offer.