AWS vs Azure in 2021 (Comparing the Cloud Computing Giants)
AWS vs Azure is a question you’ll bump into often when it comes to cloud computing. Both their cloud ecosystems offer countless benefits from superior compute power, scalability, and security to unrivaled cost-effectiveness and carbon footprint reductions.
The cloud computing industry has grown rapidly, exploding into a vast array of cloud providers, technologies, products, and services. Even a simple cloud deployment offers hundreds of options. It’s strange to say, but it’s an issue of too much choice!
To complicate things further, many providers use unique terminologies for similar offerings. The same tech will have different names, muddling like-for-like feature comparisons. Hence, you need a guide to help make difficult decisions simple.
In this article, we’re going to help you decide between the two giants of cloud computing: Amazon Web Services (AWS) and Microsoft Azure. You’ll learn about their core products, services, pricing models, and customer support structures.
Excited? Let’s get started!
Why AWS vs Azure
Every industry has its market leaders—a select few companies that rise above the rest, setting the benchmark for excellence.
If you’re exploring cloud products and services, then Google Cloud Platform, Microsoft Azure, and Amazon Web Services are the three providers who’ve become synonymous with “cloud,” with Alibaba Cloud joining the fold recently.
Before rising to prominence in the cloud market, Amazon and Microsoft were global leaders in their respective fields. Each was uniquely renowned for a history of innovation, excellence, and market dominance.
Embarking into the cloud frontier, they had the ideal technological foundation, expertise, and financial resource to develop industry-leading cloud computing platforms. Both providers have led the creation of new cloud products and services since the technology’s emergence.
Unsurprisingly, Microsoft and Amazon Web Services have again been named Leaders in Gartner’s most recent Magic Quadrant for Cloud Infrastructure and Platform Services. Both feature highest in the top-right corner of the Leaders quadrant, awarded on Ability to Execute and Completeness of Vision. AWS takes the honor of a first place, securing the top spot for the tenth consecutive year.
AWS and Microsoft Azure Dominate Cloud Market Share
According to the latest research from Canalys and Synergy Research Group, Microsoft Azure and AWS combined control more than 50% of worldwide cloud infrastructure services spend. This is a trend that seems only set to continue, with both providers seeking to further strengthen their foothold in the market through investment and innovation in their cloud platforms.
Despite the Coronavirus pandemic causing a global economic decline, the cloud computing market is defying the odds, with Gartner’s most recent forecast predicting growth of 6.3% in Worldwide public cloud revenue in 2020.
The pandemic has necessitated the need to work remotely, creating an explosion in remote working and leading to an astonishing 94% increase in the Desktop as a Service (DaaS) market. It’s similarly forcing countless organizations to begin or accelerate their digital transformation journeys to survive. Against this backdrop, you can expect AWS and Azure to continue their growth.
Regardless of whether you’re seeking IaaS, PaaS, or SaaS, you’ll discover a competitive cloud solution from Amazon Web Services and Microsoft Azure. Both providers have evolved to offer hundreds of cloud products and services, allowing them to compete across the vastness of the cloud market—their continued innovation adding new cloud solutions that create new and lucrative revenue streams.
Amazon Web Services Revenue in 2020
AWS reached an annual revenue run rate of over $40 billion at the end of 2019. Their 2019 Q4 Earnings Release reported AWS sales revenues of almost $10 billion.
As 2020 unfolded and the severity of the pandemic took hold, we began to see a significant impact on the growth rate of AWS. The Q1, Q2, and Q3 earnings release statements for 2020 showed the YoY growth rate dropping below 30%—a marked decline over the previous 3 years when growth was consistently between 40-50%.
Of course, this is not a doom and gloom scenario. Andy Jassy, the AWS CEO, kicked off the recent Amazon Web Services re:Invent Conference by announcing that the company had now surpassed $40 billion in annual revenue. As of November 2020, AWS added an incremental $10 billion in revenue in the 12 months prior—its faster ever.
Microsoft Azure Revenue in 2020
While Amazon specifies its AWS revenue, Microsoft only reports on Azure’s growth rate. This lack of specificity frustrates most cloud pundits, as we simply can’t directly compare the size of AWS and Azure.
Microsoft does, however, report on a grouping of products that Azure forms part of called “Intelligent Cloud.” When Microsoft released their Q4 2019 Earnings Release statement, they reported Azure revenue growth of 62%, with Intelligent Cloud posting revenue growth of 27% to $11.9 billion.
Against the backdrop of the pandemic, Microsoft continued to report impressive growth for 2020. Its Q1, Q2, Q3, and Q4 earnings release statements reported quarterly growth for Azure at 59%, 47%, 48%, and 50%, respectively.
As we move into 2021, Microsoft reported its commercial cloud run rate had reached $66.8 billion, with CEO Satya Nadella citing digital transformation and cloud demand as the driving force behind Microsoft’s impressive results.
Amazon Web Services vs Microsoft Azure Features Comparison
Comparing the AWS and Azure cloud platforms is no simple task. As traditional systems have moved from on-premises to cloud, both providers have expanded their service offerings to include over 25 different cloud solution categories.
Today, AWS and Microsoft Azure offer hundreds of competitive cloud solutions encompassing countless products and services. There is a wealth of choice with categories spanning compute, storage, database, security, robotics, machine learning, and even quantum technologies. To avoid getting lost in the details while comparing apples to apples, you’ll need a basic level of knowledge and understanding of the two technologies.
Thankfully, products and services from the Amazon Web Services Platform and Microsoft Azure Platform are mostly grouped under the same category headings. To help you accelerate and simplify the decision-making process, we’ve taken the time to compare the most commonly sought-after cloud products and services across several business-critical categories.
As experts in delivering our clients market-leading WordPress hosting solutions, we have first-hand experience using these types of cloud services. On top of discussing building a cloud deployment, we’ll also take a detailed look at the key considerations that accompany these services, like customer support, global infrastructure, billing, and pricing structures.
Use the navigation below to move between the sections:
- Billing and pricing
Compute resource is the foundation on which you’d build your cloud deployment. The decisions you make in this category will directly influence the speed and performance of your platform. Including the systems you run on it and the services your employees utilize. So, you must choose the right configuration for your business needs.
You also need to factor in compute cost, as it forms two-thirds of the average enterprise cloud spend, according to ParkMyCloud CEO, Jay Chapel.
Comparing Amazon Web Services vs Microsoft Azure compute capabilities, we are focused primarily on virtual machines (VMs). Forming your cloud environment’s backbone, VMs emulate physical computer systems’ functionality and power almost any workload you can think of.
Both AWS and Azure adopt a similar approach to VMs. However, as you dig into two services, you’ll discover they use different terminology for their individual compute offerings.
Amazon Web Services compute offering is known as Amazon Elastic Compute Cloud (Amazon EC2). Microsoft, on the other hand, refers to its compute product as Azure Virtual Machines. The table below maps key compute differences between AWS and Azure:
|Virtual machines||Instances||Virtual machines|
|Images||Amazon Machine Image||VM Image (both boot-disk-only and full machine)|
|VM templates||AWS Cloud Formation||Azure Resource Manager|
|Automatic instance scaling||Auto Scaling||Azure Autoscale|
|Supported VM import formats||RAW, OVA, VMDK, and VHD||VHD|
|Deployment locality||Zonal||Regional (equivalent to Cloud Platform zones)|
Virtual Machine Features
As you explore VM-instance deployments in Amazon EC2 and Azure, you’ll discover the providers share many similar—if not identical—features. These include:
- The ability to use stored disk images to create instances
- On-demand capabilities to launch and terminate instances
- Restriction-free management of your instances
- The ability to tag your instances
- A variety of available operating systems to install on your instance
Virtual Machine Access
You’ll find that both Azure and AWS adopt a similar approach to VM access of Linux and Windows machines.
For Linux machines, if you want SSH-based terminal access, Amazon EC2 and Azure both require you to include your own key SSH key. In addition to this, neither provider supports SSH browser access.
When it comes to VM access for Windows machines, Amazon EC2 and Azure support access through standard Remote Desktop Protocols (RDP).
Where they do differ slightly is in the provision of additional access pathways. Azure offers additional access to Windows machines via Microsoft PowerShell, while Amazon EC2 provides additional Windows machine access through its IPv6 address and Session Manager.
Virtual Machine Instance Types
To simplify and accelerate the process of deploying your VM setup, Azure and Amazon EC2 both offer a wide range of predefined instances.
Each VM instance type is configured with a specific virtual CPU, RAM, and network components. Whether opting for Microsoft or Amazon, there are hundreds of virtual machine types available to choose from.
Both providers build flexibility into the process, allowing you to customize your configuration. You can reconfigure core elements of the predefined instance, including the number of CPUs and available RAM, giving you the power to scale your VM resource capabilities up or down to match your organization’s unique requirements.
Both providers offer extreme high-end specifications to support the most demanding workloads. They currently max out with the following specifications:
- Amazon EC2 VMs scale up to 448 vCPUs and 24,576 GB of RAM
- Microsoft Azure VMs scale up to 416 vCPUs and 11,400 GB of RAM
To further simplify the process, both providers group VM types into categories optimized and configured for their planned usage. These VM categorizations include general-purpose, memory-optimized, compute-optimized, storage-optimized, graphics processing (GPU), and high performance.
We’ve compiled the following table, which lists the up-to-date machine types for both services as of January 2021.
|Machine Type||Amazon EC2||Azure|
|General Purpose||a1.medium – a1.metal
t4g.nano – t4g.2xlarge
t3.nano – t3.2xlarge
t3a.nano – t3a.2xlarge
t2.nano – t2.2xlarge
m6g.medium – m6gd.metal
m5.large – m5d.metal
m5a.large – m5ad.24xlarge
m5n.large – m5dn.24xlarge
m4.large – m4.16xlarge
|A1 v2 – A8m v2
B1LS – B20MS
D2a v4 – D96a v4
D2as v4 – D96as v4
D2 v4 – D64 v4
D2d v4 – D64d v4
D2ds v4 – D64ds v4
D2s v4 – D64s v4
D2 v3 – D64 v3
D2s v3 – D64s v3
D1-5 v2 – D5 v2
DS1-5 v2 – DS5 v2
DC1s v2 – DC8 v2
|Memory-optimized||r6g.medium – r6gd.metal
r5.large – r5d.metal
r5a.large – r5ad.24xlarge
r5n.large – r5dn.24xlarge
r4.large – r4.16xlarge
x1e.xlarge – x1e.32xlarge
x1.16xlarge – x1.32xlarge
z1s.large – z1d.metal
|E2 v3 – E64 v3
E2 v4 – E64 v4
E2a v4 – E96 v4
E2as v4 – E96as v4
E2d v4 – E64d v4
E2ds v4 – E64ds v4
E2s v3 – E64s v3
E2s v4 – E64s v4
D11 v2 – D15 v2
DS11 v2 – DS15 v2
G1 – G5
Gs1 – Gs5
M8ms – M128ms
M32dms v2 – M192idms v2
M32ms v2 – M192ims v2
M208s v2 – M416ms v2
S96 – S576m
S223 – S896oom
DS11-1 v2 – M128-64ms
|Compute-optimized||c6g.medium – c6gd.metal
c5.large – c5d.metal
c5a.large – c5ad.24xlarge
c5n.large – c5n.metal
c4.large – c4.8xlarge
|F2s v2 – F72s v2
F1 – F16
F1s – F16s
|Storage-optimized||i3.large – i3.metal
i3en.large – i3en.metal
d2.xlarge – d2.8xlarge
h1.2xlarge – h1.16xlarge
|L8s v2 – L80s v2
L4s – L32s
p3.2xlarge – p3db.24xlarge
p2.xlarge – p2.16xlarge
inf1.xlarge – inf1.24xlarge
g4dn.xlarge – g4dn.metal
g3s.xlarge – g3.16xlarge
f1.2xlarge – f1.16xlarge
|NC6 – NC24
NC6 Promo – NC24r Promo
NC6s v2 – NC24s v2
NC6s v3 – NC24s v3
NC4as T4 v3 – NC64as T4 v3
NP10s – NP40s
NV6 – NV24
NV12s v3 – NV48s v3
ND6s – ND24s
|High performance||n/a||H8 – H16m
H8 Promo – H16mr Promo
Virtual Machine Images
VM images accelerate your deployment by providing a pre-configured VM setup that includes an operating system, along with the supporting server and database software. Both Azure and Amazon EC2 allow the use of machine images to create new instances.
In addition to a range of proprietary ready-made image configurations, both platforms support the use of images developed by third-party vendors, which you can access through their respective platforms, AWS Marketplace and Azure Marketplace. You also have the option to create and store your own custom images for private use.
In December 2020, AWS launched Amazon Elastic Container Registry Public (ECR Public), offering the unique capability to store, manage, share, and deploy container images for anyone to discover and download globally. Proving already popular, this is no doubt a service Azure will look to replicate in the future.
Automatic Scaling of Virtual Machine Instances
Autoscaling allows you to create and remove VM instances inline with user-defined policies. You can optimize performance, scaling compute resources up or down in real-time to meet demand. This allows you to control costs, minimizing unutilized resources, so you only pay for what you need.
Both Amazon EC2 and Azure support autoscaling, implementing it in similar ways:
- AWS Auto Scaling scales instances in groups. Each group has a launch configuration to create new instances and uses your chosen scaling plan to manage the creation and removal of instances.
- Azure Autoscale has a VM scale set in which instances are scaled. Instances are created or removed inline with your chosen scaling plan, referred to as an autoscaling policy.
Between the platforms, there are three autoscaling plans available: manual, dynamic, and scheduled. Amazon Auto Scaling supports all three, while Azure Autoscale only supports dynamic and scheduled autoscaling. Each plan can be defined as follows:
- Manual: You can manually instruct the creation and removal of instances.
- Scheduled: Instances can be scaled up or down based on a predefined schedule.
- Dynamic: You can create policies to scale instances based on specific metrics such as CPU utilization or message queue length.
Temporary Virtual Machine Instances
Temporary instances are VMs that run on the cloud provider’s unused capacity. These VMs are unpredictably available, so you can reallocate their resources at any given moment. As a result, they are available at highly discounted rates, letting you unlock the power of the cloud for less.
Temporary instances are ideal for:
- Workloads that can be interrupted without losing work
- Low priority jobs that are not time-sensitive
- Workloads that benefit from increased compute power when available, such as for rendering video
Both Amazon and Microsoft cloud platforms support temporary instances with similar naming conventions. The AWS service for temporary instances is Spot Instances, while the Azure service is Spot Virtual Machines (Spot VMs).
Both Azure and AWS temporary machines share a set of features that include:
- The ability to control temporary instances when they are running
- Limiting the instance types and machine images available, compared with on-demand instances
- Accessing the same performance as on-demand instances when temporary instances are running
Temporary instances on AWS or Azure are available at a discounted rate of up to 90% compared with standard pay-as-you-go, on-demand pricing. It’s certainly worth investigating further.
Comparing VM machine performance between Azure and AWS is no easy task. We can’t simply declare that one of these cloud providers is better than the other. There are hundreds of comparable VM instances available between AWS and Azure, with the performance scale tipping one of two ways, depending on the comparison.
A recent study from Cockroach Labs compared AWS vs Azure vs GCP CPU performance across a range of single-core and 16-core VMs. GCP came out on top in the single-core category, with performance 10% higher than AWS, with Azure coming in last. When comparing 16-core VMs, AWS came out on top with the fastest iterations per second. GCP was second, and Azure took last again.
Azure and AWS cloud platforms are built on a global cloud infrastructure of hundreds of data centers interconnected through hundreds of thousands of lit fiber optic and undersea cable systems. Each is renowned for delivering state-of-art networking services that offer high-speed performance, high availability, strong security, and global coverage.
This section will explore in detail the core networking products and services available from Microsoft and Amazon. The following table gives you a high-level view of AWS and Azure’s comparative networking products before the deep dive.
|Product||Amazon Web Services||Microsoft Azure|
|CDN||Amazon CloudFront||Azure CDN|
|Dedicated Interconnection||AWS Direct Connect||ExpressRoute|
|DNS||AWS Route 53||Azure DNS|
|Load Balancing||Elastic Load Balancing||Azure Load Balancer|
|Virtual Networks||Amazon Virtual Private Cloud||Azure VNet|
Data Center Network Locations
Both AWS and Azure offer a global network of interconnected data centers that span multiple continents, regions, countries, and locations. Both platforms provide comprehensive network redundancy, high fault tolerance, and low latency.
Amazon Web Services Network Locations
At last count, AWS cloud network locations are available in 24 regions, 77 zones, 210 network edge locations, and 245 countries and territories. As a result, AWS has the largest global cloud infrastructure footprint of any provider.
This gives AWS a distinct edge when it comes to network latency. On average, you and your customers will be closer to an AWS data center. This geographic proximity means data has less distance to travel, reducing latency times.
Amazon has ambitious plans for the continued expansion of its global infrastructure by launching additional data centers. The locations of these centers include Hyderabad (India), Jakarta (Indonesia), Osaka (Japan), Madrid (Spain), and Zurich (Switzerland).
Azure Network Locations
Though not as expansive, Azure cloud network locations are available in 60+ regions and 170 network points of presence (PoP).
Microsoft also has ambitious expansion plans for its Azure global network. Plans for new Microsoft data centers include Austria, Chile, Denmark, Greece, Israel, Italy, Mexico, New Zealand, Poland, Qatar, Spain, Sweden, Taiwan, and the US (Arizona).
Content Delivery Network (CDN)
Using a Content Delivery Network (CDN) will ensure your users get the fastest access to your content, wherever they are based. AWS and Azure each offer a CDN solution: Amazon CloudFront and Azure CDN, respectively.
By distributing and delivering content locally, you’ll benefit from quicker load times, reduced bandwidth strain, and improved responsiveness of your applications and websites. Each provider delivers this service by replicating and hosting your content across their distributed global network of interconnected data centers.
Both CDN solutions deliver enhanced security to protect network and transport layer DDoS attacks, offering improved resilience and preventing service loss.
CloudFront bolsters security by offering deep integration with AWS security products, including AWS Shield, AWS Web Application Firewall, and Route 53. At the same time, Azure CDN delivers enhanced protection through customized third-party solutions like Azure CDN from Verizon and Azure CDN from Akamai. Microsoft provides a detailed feature comparison between these DCN services via its product documentation platform.
A DNS service is used to convert human-readable domain names into the IP addresses used by servers to communicate.
Where Amazon Route 53 has an edge is in the provision of three distinct features, not yet supported by Azure DNS:
- Geography-based routing, letting you restrict content to specific geographic locations
- Latency-based routing, or the ability to direct traffic according to latency levels measured by the DNS services
- DNSSEC signing, providing validation that a DNS response came from Amazon Route 53 and hasn’t been tampered with
The table below outlines a list of features mapped across both services:
|Feature||Amazon Route 53||Azure DNS|
|Zone||Hosted Zone||DNS Zone|
|Support for most DNS record types||Yes||Yes|
|Private Zones / Split Horizon||Yes||Yes|
A load balancing service, when correctly configured, automatically distributes traffic across multiple VM instances to improve the performance, fault tolerance, and availability of your applications.
Both Microsoft and Amazon offer comparable load balancing products that you can configure to meet your needs.
AWS Load Balancing
The load balancing solution for AWS is called Elastic Load Balancing (ELB). It comes complete with the following features and functionality:
- Load balance both internal and external traffic to AWS instances
- Increase availability by distributing traffic to VM instances across multiple availability zones in a specified AWS region
- Target instances receive regular health checks, with traffic redirected when an instance is deemed unhealthy
- ELB integration with AWS Auto Scaling, which automates the creation and removal of VM instances, allowing you to optimize compute resource in line with demand automatically
- Content-based routing and SSL available through an Application Load Balancer
- High throughput and low latency Layer 4 connections through a Network Load Balancer
- Ability to configure your load balancer to work with IPv6 addresses
For a closer look and a more detailed feature comparison, check out AWS’s Elastic Load Balancing documentation.
Azure Load Balancing
Microsoft’s load balancing solution on Azure is called Azure Load Balancer. It offers the following features and functionality:
- Load balancing for internal and external traffic to Azure virtual machines
- Increase in availability by distributing traffic across VM instance within and across zones
- Use of health probes to monitor VM performance and redirect traffic in the event an instance becomes unhealthy
- Support for load balancing your services across multiple ports, multiple IP addresses, or both
- Use of PowerShell and Azure Cloud Shell to support load balancing of IPv6
Check out Microsoft’s Load Balancer documentation for a more detailed look at their product offerings.
Private Connectivity to Other Networks
Both AWS and Azure offer multiple services to support private connectivity outside their network, including your on-premises environment.
The table below provides a high-level comparison of AWS and Azure private connectivity services:
|Feature||Amazon Web Services||Microsoft Azure|
|Virtual Private Network||Amazon CloudFront||Azure VPN Gateways|
|Private connectivity to a VPC||AWS Direct Connect||VNet peering|
|Dedicated private connection through a carrier partner||AWS Direct Connect||ExpressRoute|
|Dedicated public connection through a carrier partner||AWS Direct Connect||ExpressRoute|
|CDN connectivity||Amazon CloudFront||Azure CDN|
AWS Private Connectivity
AWS Direct Connect is the primary service for delivering private connectivity to the AWS platform. AWS Direct Connect lets you create a dedicated network connection between your network and one of the hundreds of AWS Direct Connect locations.
Establishing this connection reduces your bandwidth costs by transferring data directly to AWS, minimizing data usage through your existing ISP. You also benefit from a reduced AWS Direct Connect data transfer rate rather than data transfer rates.
Using Direct Connect to link your on-premise network and Amazon VPC will create a private, high-bandwidth network. The service’s elastic nature allows you to scale your connection from 1 Gbps to 10 Gbps to meet your business’s unique needs.
Azure Private Connectivity
Azure ExpressRoute is Microsoft’s primary private connectivity service for the Azure platform. Like AWS, you can create a dedicated network connection between your on-premises network and the Microsoft global network.
You’ll access similar bandwidth cost reductions through reduced ISP usage and reduced data transfer rates available through Azure ExpressRoute.
ExpressRoute appears to have the edge over Direct Connect when it comes to connection speeds, with Microsoft offering a bandwidth of up to 100 Gbps supported through their direct connection.
Latency is a critical factor when comparing cloud networks. It’s the time delay between your client’s request and the cloud service provider’s response.
Cockroach Labs 2021 Cloud Report compared network latency of Azure, AWS, and GCP. Their research showed AWS performed best with the lowest network latency overall, closely followed by Azure and, finally, GCP.
While the research is a robust and informative measure, you’ll still need to factor in your own unique requirements. As the above study highlights, latency varied in the research based on the VM instances’ physical placement.
Your geographic proximity to the cloud provider’s data center will significantly influence latency, so you must explore this factor when comparing AWS and Azure.
This section takes a more detailed look at the different storage types and services available on the Azure and AWS platforms.
Comparing cloud storage options between the Microsoft and Amazon cloud platforms is split into five distinct storage service types:
- Distributed object storage: Redundant key-value stores in which you can store data objects
- Block storage: Virtual disk volumes you can attach to a VM instance
- File storage: Designed for file-server-based storage
- Cool storage: Designed for the storage of backups
- Cold (archival) storage: Designed for storing archival data inline with compliance or analysis purposes
Understanding the differences between storage and disk types is vital in choosing the right service for your business. These are critically important decisions as they will directly influence your applications and systems’ speed and performance using cloud storage services.
Distributed Object Storage
Distributed object storage is the process of storing data as objects, otherwise referred to as blobs. This process lets you store, scale, secure, and access volumes of object data used across multiple scenarios, including websites, cloud-native mobile apps, backups, archival, and big-data analytics.
Amazon Simple Storage Service (S3) and Azure Blob Storage are the competing object storage services. Despite a variation in terminology used to describe their services, they offer similar functionality.
Each object is stored in a set unit, and has a unique key and an associated metadata record containing pertinent information: object size, date of last modification, media type, etc.
Both providers offer a similar feature set for their object storage services, which include:
- Capability to host web and static media content
- Object versioning, whereby objects are stored as multiple distinct versions, preventing data from being overwritten and the resulting loss of data
- Object lifecycle management for automation of the migration and deletion of objects using preset, user-specified lifecycle policies
- Update notifications, wherein you configure notifications to be triggered following object creation, updates, or deletion
- Service Level Agreement (SLA): Amazon S3 and Azure Blob Storage have SLAs with uptime guarantees offering a tiered refund amount once uptime drops below 99.9%.
Below is a table comparing the features of distributed object storage between Azure Blob Storage and Amazon S3:
|Feature||Amazon S3||Azure Blob Storage|
|Unit of Deployment||Bucket||Container|
|Deployment identifier||Globally unique key||Account-level unique key|
|File system emulation||Limited||Limited|
|Object versioning||Yes||Manual, per-object snapshotting|
|Object lifecycle management||Yes||Yes (through lifecycle rules or Azure Automation)|
|Update notifications||Event notifications||Event notifications|
|Service classes||Standard, Standard-Infrequent Access, One Zone-Infrequent Access, Amazon Glacier||Redundancy levels: LRS, ZRS, GRS, RA-GRS
Tiers: Hot, Cool, Archive
|Deployment locality||Regional||Zonal and regional|
|Pricing||Priced by the amount of data stored per month, network egress, and number of common API requests||Priced by volume of data stored per month, quantity and types of operations performed, data transfer costs, and any data redundancy options selected|
Block storage is where a virtual disk is attached to a cloud-based virtual machine.
Both Amazon and Microsoft deliver block storage services that fully integrate with their respective cloud platforms. Each offers a range of configurable block storage types that you can scale to meet your business’s unique requirements.
Network-attached disks are volumes connected to your VM instance via your cloud provider’s network. This delivers cloud-native benefits such as built-in redundancy, disk snapshots, and the seamless attachment and detachment of disk volumes.
Here’s how Microsoft and Amazon compare on features for network-attached disks in their block storage services:
|Feature||Amazon EBS||Azure VHDs|
|Volume types||EBS Provisioned IOPS SSD, EBS General Purpose SSD, Throughput Optimized HDD, Cold HDD||Ultra Disk Storage, Premium SSD, Standard SSD, Standard HDD|
|Volume attachment||A single volume can be attached to up to 16 instances, each with read-write permissions to the shared volume||A single volume can be attached to up to 10 instances and can only be configured with a read-only status|
|Maximum volume size||16 TiB||64 TiB|
|Disk encryption||Encrypted by default||Encrypted by default|
Let’s take a closer look at some key features for network-attached disks:
Volume Attachment and Detachment
Until late 2020, Google Cloud was the only provider to offer the capability for a virtual disk volume to be attached to multiple instances, a unique feature that would allow you to deploy a new—or migrate an existing—clustered application.
Seeing the inherent benefits of this feature, Amazon and Microsoft have now released competitive functionality on their respective cloud platforms: EBS Multi-Attach for AWS and Shared Disks for Azure.
Amazon EBS and Azure both allow you to create and store snapshots of your disk volumes. You can use these snapshots to create new volumes at a later date. You can use the Snapshots feature to back up critical workloads and data for failover in the event of system failures.
Snapshots in AWS and Azure are incremental. The first snapshot replicates the entire disk volume, with subsequent snapshots only creating copies of the blocks that have changed since the previous snapshots.
Locally Attached Disks
Locally attached disks are volumes that have been directly connected to the physical machine running your instance.
Attaching a disk locally reduces latency and delivers higher throughput for increased performance, meaning data loads faster, improving user experience. The downside is that if a locally attached disk fails, there is no redundancy in place.
Let’s take a high-level look at how AWS and Azure features compare for locally attached disks:
|Block Storage||Amazon EC2||Azure|
|Service name||Instance store||Local SSD|
|Volume attachment||Tied to instance type||Tied to instance type|
|Attached volumes per instance||Varies by instance type – up to 24||Varies by instance type|
|Storage capacity||Varies by instance type – up to 2500 GB per volume||Varies by instance type – up to 2400 GB per volume|
Delivering a fully managed service, AWS and Azure let you quickly create and configure file share systems with ease, free from the burden of deploying and managing the underlying infrastructure.
Security is ensured through encryption of data by default, both at rest and in transit. Automation features allow you to scale your file system storage capacity up or down, offering cost efficiency and optimized performance in line with demand.
Let’s take a closer look at some key feature differences in AWS and Azure file storage services:
Network File System Protocols
Both Amazon EFS and Azure Files support NFS v4, the newer and higher-performing Network File System Protocol. However, NFS v4.1 is only supported in preview for Azure Files as of September 2020, while that same functionality has been available through Amazon EFS since early 2017.
Input/Output Operations Per Second (IOPS)
When we compare read operations per second, Amazon EFS has a significant edge over Azure Files.
In April 2020, Amazon announced a 400% increase to read operations for General Purpose mode file systems through the Max I/O performance mode, supporting 500,000 IOPS and 10 GB/s of throughput.
According to Azure file share targets, Azure File only supports 100,000 IOPS and 10 GB/s of throughput via its Premium file share service, making Amazon EFS read operations 5x faster than Azure File at the top end of service delivery. Of course, this performance difference is only relevant to a select few of you who plan to house the largest databases with the most demanding workloads.
For the average enterprise, your file storage demands can be equally met by both AWS and Azure.
Cool storage is designed for data stored for long periods of time, which is rarely accessed. Cool storage is typically used for database and file backups.
AWS offers cool storage through Amazon S3 storage classes. There are two classes available for cool storage:
- Amazon S3 Standard-Infrequent Access (S3 Standard-IA): For data accessed less frequently, but requires rapid access when needed.
- Amazon S3 One Zone-Infrequent Access (S3 One Zone-IA): Offering the same service as S3 Standard-IA, but in only one availability region. This service costs 20% less, ideal if you want a lower-cost option and are not concerned about reduced availability and resilience.
Azure offers cool storage through Azure Blob Storage access tiers. There is only one relevant tier for cool storage:
- Azure Blob Storage Cool: Optimized for storing infrequently accessed data for at least 30 days.
Archival storage is designed for data that is stored for long-term preservation, which is rarely accessed. This service is low-cost in comparison with other storage types due to the reduced workload requirements.
Archival storage is typically designed primarily for those of you working in highly regulated industries—Financial Services, Healthcare, and the Public Sectors—where you’re required to retain datasets for 7–10 years to meet regulatory compliance.
AWS offers archival storage through Amazon S3 storage classes. There are two classes available for archival storage:
- Amazon S3 Glacier (S3 Glacier): Supporting a range of time-related retrieval options and availability—for data accessed perhaps once a year.
- Amazon S3 Glacier Deep Archive (S3 Glacier Deep Archive): The lowest-cost option designed primarily for data stored over the years for regulatory compliance.
Azure offers archival storage through Azure Blob Storage access tiers. There is only one relevant tier for archival storage:
- Azure Blob Storage Archive: Optimized for storing infrequently accessed data possessing flexible latency requirements for at least 180 days.
Security will be among your primary considerations when choosing a cloud provider.
When comparing cloud security between AWS and Azure, look closely at the core security pillars that will combine to protect your cloud-based applications, data, infrastructure, and systems. This means exploring controls, policies, processes, and technologies that define your security stance.
Amazon Web Services and Microsoft Azure are renowned for their robust security stance and the implementation of cutting-edge security. Both continually invest in cybersecurity research and development to deliver a secure platform, resilient to the ever-evolving and escalating threat of the technology landscape.
Amazon and Microsoft deliver cloud security through a combination of three practices:
- Security of their cloud: Protecting you by default through their underlying cloud platform infrastructure’s built-in security capabilities.
- Security in the cloud: Allowing you to enhance your cloud deployment’s protection by providing security products and service add-ons available within their cloud platform.
- Security anywhere: Expand security beyond their cloud platform to protect your assets regardless of location through encryption protocols.
Let’s take a closer look at AWS cloud security vs Azure cloud security and the core features and functionality you should consider as part of that comparison.
Data protection and compliance continue to rise as both governments and industries apply increased control over the storage, transfer, and sharing of information. Whether you operate in a highly regulated industry, a specific geographic location, or both, compliance with your cloud platform will be a key security consideration.
Each offers comprehensive compliance programs to cover certification, law, industry-specific regulation, security frameworks, and privacy. There is a distinct crossover between both cloud providers as each competes to serve the same consumer base.
With GDPR continuing to be a major consideration and concern for many organizations, both AWS and Azure have developed resource centers. As well as highlighting how their respective cloud platforms are GDPR-compliant, both offer a wealth of documentation and guidance on best practices to ensure your organization meets GDPR requirements.
If it’s purely a numbers game, Microsoft appears to have the edge over Amazon. The Azure compliance program aligns to 91 compliance standards versus the AWS compliance program, aligning only to 75 standards.
For you, choosing a cloud provider based on compliance will only become relevant if a specific industry or government standard is applied to your business. Only then is it likely to be a decisive factor in your choice of platforms.
Encryption is critical to the protection of your data, regardless of its location. The practice of encoding your data ensures it is near impossible to decipher without a decryption key, maintaining security and secrecy.
Both AWS and Azure cloud platforms offer your data encryption by default, both at rest and in transit. Data is encrypted using 256-bit AES, one of the strongest encryption ciphers available.
AWS Key Management Service (KMS) and Azure Key Vault are their competing encryption services. Both let you centrally create and manage the keys used to encrypt and digitally sign your data, allowing you to manage encryption across their full range of cloud services.
Both providers’ encryption services meet the Federal Information Process Standard 140-2 (FIPS 140-2), which validates that their cryptographic modules meet well-defined security standards.
A firewall is the first line of defense for your infrastructure. It is responsible for protecting your network against unwanted intrusion.
Both Microsoft and Amazon ringfence their cloud platform with state-of-the-art firewalls, providing you with a base level of protection. In addition to this, both cloud providers offer a range of competitive firewall-as-a-service products to enhance your security stance.
Outlined below are the different firewall services available along with the competing products:
Centrally configure and manage firewall rules across all your accounts, applications, instances, and locations.
Web Application Firewalls
Rapidly deploy a firewall service to protect your web applications from common web exploits that can impact your availability, compromise security, or consume excessive compute resources. Control how traffic reaches your applications and create rules to block common attacks, like SQL injection and cross-site scripting.
Virtual Private Network Firewalls
Access a managed firewall service, allowing you to deploy essential network protection across your cloud-based VPNs quickly. Centralize the creation, enforcement, and logging of your application and network connectivity policies.
Identity and Access Management
Identity Access Management is critical in preventing private and sensitive information from being accessed by the wrong people.
It is the process of applying user-based security to control who has access to your applications, files, folders, systems, and what that user can do if they have been given access. This typically controls their right to access, create, edit, and delete files, services, or settings.
Amazon and Microsoft offer a range of services allowing you to implement Identity Access Management in your cloud deployment.
Identity and Access Management Console
A central console through which you can manage and secure identities. Create and manage users and groups, applying permissions to allow or prevent their access to cloud-based resources.
Mobile and Web Application Control
Implement identity and access management to control user sign-up, sign-in, and control access to your customer-facing mobile and web applications.
Use single sign-on (SSO) access to manage and control user access to multiple accounts and applications on-premises and in the cloud, delivering ease of access for an improved user experience and your employees’ heightened productivity.
Shared Responsibility Model
Implementing security in the cloud is a shared responsibility.
Understanding where responsibility ends and begins between you and your cloud provider is critical in maintaining a strong security stance. A misunderstanding here can create vulnerabilities and gaps in your security that would otherwise be easily avoidable.
Amazon Web Services and Azure provide clear documentation on their shared responsibility model, giving you the information you need to understand where security responsibility sits. Below, you can see a visual representation of shared responsibility for each cloud provider:
AWS Shared Responsibility Model
Azure Shared Responsibility Model
When managing your cloud deployment, you will encounter situations where you lack the knowledge to perform certain tasks or troubleshoot issues. When this happens, you need a cloud provider who can support you with effective guidance and additional support, so you can quickly and easily resolve these challenges.
AWS and Azure are widely recognized for their self-service support. This includes extensive repositories of technical documentation and thriving online communities spanning countless blogs, groups, and discussion forums.
Here you’ll encounter hundreds of thousands of cloud experts openly sharing their knowledge and expertise. You are free to browse an ever-growing ecosystem of topics, tutorials, discussions, and virtual or in-person meetups.
Below, you can find links to these invaluable self-service support systems:
You cannot resolve all situations through self-service. Eventually, you’ll encounter a situation where you need the immediate hands-on support of an expert. When this situation arrives, it’s best to have an official support solution in place, direct from your cloud provider.
Both AWS and Azure offer basic cloud support plans, along with a range of paid premium plans. If you’re considering a premium plan, research and understand what’s included along with the associated fees, ensuring you pick a plan you need at a price you can afford.
AWS Cloud Support Plans
AWS has four available support plans split between free and premium tiers. Premium support is divided across three tiers: Developer, Business, and Enterprise
Pricing starts from $29/month or 3% of AWS usage and scales upwards over $15,000/month. Business and Enterprise pricing is calculated at a percentage of AWS usage that decreases across brackets, as seen below:
- 10% of the first $150k
- 7% from $150k to $500k
- 5% from $500k to $1 million
- 3% from $1 million+
Each support tier increase adds to your available options, including:
- Best practice checks
- Additional communication channels
- 24/7 availability
- Issue response times inside 15 minutes for business-critical outages
- Support API
- Training resources
- Access to a Technical Account Manager
Premium plans are fully customizable, letting you choose which products and services you want to add premium support to and giving you greater control over your costs, along with a support service that reflects your requirements.
Azure Cloud Support Plans
Microsoft offers five Azure support plans: Basic, Developer, Standard, Professional Direct, and Premier.
Basic, Developer, and Standard support plans are role-based and offer plans ranging from free to $100/month per user. Each level increase adds additional layers of support, including:
- More support types
- More communication channels
- Faster response times
- General architecture support
Professional Direct support is the Goldilocks package, positioning its appeal in support and pricing between Role-based support and Premier support. It comes with a price tag of $1000/month per user and is aimed at organizations with a business-critical dependence on Azure.
Professional Direct is a step above Standard support with additional services, including:
- Response times under 1 hour
- Architecture support
- Operations support
- A dedicated Account Manager
Premier support is the highest level of support – you’ll need to contact Microsoft for pricing. It’s designed for organizations with a business-critical dependence on Azure across multiple products.
Premier support is another step up from Professional Direct, with additional services that include:
- Response times of 15 minutes
- Customizable architecture and operational support
- On-demand training
- A designated Technical Account Manager
- Launch support for new product and service deployments
At Kinsta, we understand the importance of first-class expert support. That’s why our entire support team is made up of highly skilled WordPress and Linux engineers. We provide you with access to experts who are active contributors to the WordPress core, open source projects, and even plugins development.
We believe in expert support for all. You’ll get the same level of premium support whether you’re an SME or a Fortune 500 company, making our support service second to none.
Billing and Pricing
As with any product you purchase, price is always one of the most important considerations. Unfortunately, getting an accurate pricing comparison for cloud deployments between providers will be one of the most challenging and time-consuming aspects of the decision process.
Each provider’s billing approach and usage discounts—plus the almost limitless combinations of products and services create layers of complexity, which most professionals fail to navigate successfully.
To help you understand the challenge of creating an accurate pricing comparison, here’s a small selection of variables you can control to influence the pricing of your cloud deployment:
- Virtual machines: Number of instances, RAM requirements, number of CPUs, reserved or temporary instances.
- Storage disks: Storage amount required, data types, redundancy requirements, network-attached, or locally-attached.
- Subscription model: Purchasing by the second, minute, hour, day, month, or year.
- Support: Which tier you opt for, whether you customize your support, your average monthly cloud spend
- Payment model: Whether you’re selecting a pay-as-you-go service, reserved instance, or long-term committed use contract
- Location: Datacenter location also influences pricing
The complexity of your pricing comparison increases exponentially with each additional product or service you include. The prospect is only made more challenging by manually deciphering the comparable technologies between each cloud provider.
But all is not lost! We’ve researched and uncovered a series of tools and data to guide you toward formulating your own personalized AWS vs Azure pricing comparison.
AWS vs Azure Cloud Pricing Comparison
Each provider offers its own unique pricing mechanism and a range of configurable options to influence the overall cost. Even a simple cloud deployment of a single VM instance with attached storage will come with thousands of product configurations and pricing iterations to choose from.
Understanding your business needs and the associated cloud products you need will allow you to focus on this situation. Only then can you narrow your options and begin to form your cloud pricing comparisons.
Cloud Pricing Calculators
To help you navigate and form an accurate pricing comparison, both Amazon and Microsoft have created comprehensive cloud pricing calculators, putting every product, configurable option, and the associated price at your fingertips.
AWS vs Azure Compute Pricing Comparison
To provide you with greater insight into how AWS and Azure compare in cloud pricing, we’ve conducted our own research.
We’ve chosen to focus specifically on comparing the costs of VMs from Azure and Amazon EC2. According to Gartner, this decision is based on the fact that compute resources make up two-thirds of the total spend for an average cloud deployment.
Configuration for Cloud Pricing Comparison
To create an apples-to-apples comparison between Azure and AWS VMs, we’ve chosen the same configuration options across CPUs, operating systems, and regions:
- Region: US East – North Virginia (AWS – US East, Azure – East US)
- Operating System: Linux
- vCPUs/Cores: 4
We’ve then selected VM instances with comparable RAM specifications across the following machine usage types:
When making your own configuration, make time to explore the options. You’ll quickly discover switching between the different variables across instance type, region, operating system, and CPUs can significantly alter your price estimate.
For our pricing comparison, we’ve chosen the following instances:
|Instance Type||Amazon EC2||AWS RAM
|Azure VM||Azure RAM
The pay-as-you-go pricing offers you a flexible, on-demand approach to the consumption of cloud resources. Ideally suited to organizations with intermittent cloud usage, this option allows you to add and remove cloud resources in line with demand. However, this flexibility comes at a cost, with pay-as-you-go pricing models having the highest price per hour:
|Instance Type||Amazon EC2|| AWS Price
|Azure VM||Azure Price
When comparing AWS vs Azure VM pricing, Amazon EC2 is the clear winner across general-purpose, compute-optimized, and memory-optimized instance types. AWS is a minimum of 20% cheaper across all three categories.
If you’re only interested in short-term intermittent compute resources and have workloads resistant to interruption, you should consider exploring temporary instances. Known as Spot Instances on AWS and Spot Virtual Machines on Azure, temporary instances are up to 90% cheaper than on-demand, pay-as-you-go pricing.
Long-Term Commitment Plans
If you’re planning a long-term cloud deployment, then long-term commitment plans with your cloud provider will offer a significant cost saving over the pay-as-you-go models.
Both AWS and Azure offer long-term commitment plans, which they refer to as reserved instances, where you can choose from two upfront commitments: one year or three years. Reserved instances are rewarded a cost savings of up to 72% over on-demand, pay-as-you-go pricing.
AWS offers greater flexibility for its reserved instance plans as compared to Azure. With Amazon EC2, you have the choice between two plans:
- Convertible instances: The ability to switch to a newer VM instance when it becomes available during your long-term commitment.
- Non-convertible instances: You commit to using the same VM instance for the duration of your long-term commitment.
Non-convertible instances—where you pay everything up front—offer the greatest level of discount as you sacrifice flexibility in your deployment.
When comparing reserved instances for AWS, we’ve chosen a 100% upfront payment and non-convertible instances to unlock the highest discount available. Paying upfront with Azure appears not to influence your discount percentage, at least according to the Azure pricing calculator.
One-Year Reserved Instance
|Instance Type||Amazon EC2|| AWS Price
|Azure VM||Azure Price
Comparing pricing for one-year reserved instances, AWS is the clear winner across all three instance types.
AWS is 20% cheaper than Azure across general-purpose and memory-optimized instance types, and an impressive 40% cheaper on compute-optimized instance types.
Both cloud providers offer an average of 40% cost savings for one-year reserved instances over on-demand, pay-as-you-go-pricing.
Three-Year Reserved Instance
|Instance Type||Amazon EC2|| AWS Price
|Azure VM||Azure Price
When comparing AWS to Azure pricing for three-year reserved instances, AWS continues to offer the most competitive pricing and is again the cheapest option across all three instance types.
Across general purpose and memory-optimized instance types, AWS is 20% cheaper than Azure. The price gap widens for compute-optimized instance types, with AWS being 35% cheaper than its Azure equivalent.
When it comes to three-year reserved instances, AWS and Azure offer an impressive average cost saving of 60% versus on-demand, pay-as-you-go-pricing.
A free trial is ideal for testing the cloud provider’s services without the need to make a financial commitment.
Both AWS and Azure offer free trials on a range of their core cloud services, giving you a predefined resource amount over a set period of time – perfect for testing cloud services.
In addition to this, both cloud providers also offer “always free” cloud services—ideal if you have meager cloud usage requirements and you’re not concerned with operations being interrupted. Let’s take a closer look.
AWS Free Tier
With the AWS Free Tier, you’ll have free access to 85 different AWS cloud products and services across compute, storage, database, IoT, AI, and many more categories.
The AWS Free Tier is split into three types:
- Always free: A free offer that never expires and is available to all AWS customers
- 12 months free: Available free over the first 12-months from your initial sign-up to AWS
- Trials: Free for a short-term following the activation of a particular service
If you’re exploring AWS for the first time, we recommend you inspect AWS compute and storage options included free for 12 months following sign-up. These are the services that will form the foundation of your cloud deployment:
- Compute: Amazon EC2 access for 750 hours a month with a t2. or t3. micro instance
- Storage: Amazon S3 standard storage of 5 GB per month with 20,000 read 2,000 write operations
Of course, there’s a range of restrictions, so make sure you check the AWS Free Tier FAQs to find out about key information like moving from free to pay-as-you-go, regional restrictions, and more.
Azure Free Account
Creating an Azure free account will unlock free access to 54 Azure cloud products and services across compute, storage, database, security, AI, and many more categories.
The Azure free account gives you two types of access:
- 12 months free: Access to 25 cloud products for 12 months only
- Always free: Access to 29 cloud products that always remain free of charge
In addition to your free access, you’ll also receive $200 of credit to spend during the first 30 days after sign-up.
If you’re trialing the Azure cloud platform for the first time, we’ll echo our sentiments with AWS and suggest trialing the Azure compute and storage services, which are free for the first 12 months:
- Compute: Azure access for 750 hours a month with a B1S Standard tier instance with either a Linux or Windows OS
- Storage: Azure Blob of 5 GB per month with 20,000 read 10,000 write operations
As with AWS, ensure you explore the Azure Free Account FAQs, as there’s a range of restrictions and considerations you’ll need to be aware of.
Is AWS Cheaper than Azure?
Working out which cloud provider offers the most competitive price is no easy task.
When it comes to compute pricing, based on our extensive research, we can confidently say AWS is cheaper than Azure on average—a fact that is consistent across all pay-as-you-go and reserved instance price plans.
But that’s just one aspect of your cloud deployment. Looking at the bigger picture, excuse the pun, things become clouded. Our wider research into comparative cloud products, services, and pricing models shows the balance can shift between AWS and Azure.
A study from NetApp comparing storage pricing definitively shows Azure to be the cheapest option for Object Storage. However, for Block Storage pricing, the balance shifts, and AWS becomes the cheaper option.
Upon reflection, deciding whether AWS or Azure is the cheapest cloud provider is an answer that remains unique to your organization, defined by your location, cloud deployment configuration, resource requirements, chosen pricing plan, and the other countless variables that will influence your final cost.
In comparing Amazon Web Services to Azure, we’ve sought to answer the question of which cloud provider is best overall. At the end of the journey, our research heavily favors AWS as the clear winner.
Stepping back and looking at the big picture, several key factors make this a fairly easy decision:
- AWS has consistently been hailed as the leading provider in Gartner’s Magic Quadrant
- Industry-wide research shows Amazon with a larger cloud market share than Azure
- AWS offers more cloud products and services than Azure
- Amazon’s cloud network is bigger, with more points of presence across the world
- AWS beat Azure in Cockroach Labs’ independent compute, network, and storage performance research across the board
- AWS is cheaper than Azure for compute pricing, which forms the backbone of cloud deployments
However, this research has taught us that this answer can change based on each business’s unique requirements. So while AWS might look better overall, your own research for your business may lead you to decide on Azure as the best choice for you.
In the end, the fact remains that both Azure and AWS are cutting-edge providers with the very best cloud platform, products, and services to offer. Wherever you end up, you’ll unlock a wealth of cloud benefits.
If you enjoyed this article, then you’ll love Kinsta’s WordPress hosting platform. Turbocharge your website and get 24/7 support from our veteran WordPress team. Our Google Cloud powered infrastructure focuses on auto-scaling, performance, and security. Let us show you the Kinsta difference! Check out our plans