Menu
Microsoft vs Amazon clouds: Five key differences

Microsoft vs Amazon clouds: Five key differences

2. Application support

While both Azure and EC2 heavily market applications that ISVs make available through their services, EC2 tends to attract resource-intensive software typically classified as enterprise applications, while Azure tends to feature applications that typically serve LAN- or workgroup-sized user groups.

Both platforms are designed to be easily resizable and are being developed quickly; for now at least, that minimizes but doesn't eliminate the difference between IaaS and PaaS, Golden says.

3. Application portability

Because it's a Microsoft environment and supports commonly used x86-based computing technologies such as .NET and Visual Studio, Azure is more accessible to developers accustomed to those environments, according to Gary Chen, IDC server virtualization analyst.

The benefit and drawback to that is that Azure only offers Microsoft operating systems, which makes things simpler for Windows developers, but limits the applications they can build on or port to Azure, Hackett says.

The new Windows Server capability is designed to make applications even more portable between data center and cloud, but only when both are closely equivalent Windows Server environments.

Amazon offers Linux, Mac OS X or Windows support, but has some specific storage requirements and also requires developers to take into account the amount of bandwidth and computing time that the application will require.

4. Scalability

Both Azure and EC2 are designed to be scalable, but the way they achieve this is different.

On Azure users have the choice of scaling up the number of VMs on which an application runs, or by increasing the power of the virtual machine they use, according to .NET development consultant Jason Haley. It offers storage as varying blocks of Binary Large Object (BLOB) service and virtual NTFS drives.

Pay-as-you-go on Azure means defining the number of VM nodes that an application requires and the ability to change that number using a management console.

EC2 is far more flexible now than two or three years ago, Golden says. It still scales in blocks called Elastic Compute Units (ECU) that include a specific number of Amazon Machine Images (AMI) and storage volumes called Elastic Block Stores (EBS).

ECUs include varying amounts of storage, I/O, memory, performance equating to specific 32- or 64-bit processors, and a variety of networking and higher-level systems or application-management services.

At one time Amazon offered only small, medium and large-sized resource blocks and price points. Then in September, it introduced micro instances that require as little as 613MB of memory, and cost as little as two cents per hour of use. An Amazon feature called autoscaling automates changes in resource levels to respond to demand.

5. Price

Because Amazon's service is designed to be more comprehensive, its pricing is more complex than Microsoft's, though neither pricing plan is simple.

Amazon's service is reputed to cost more in general, but developers and analysts also complain about the rigidity and surprisingly high cost of Microsoft Azure, often finding EC2 cost-effective even for small projects when comparing the two. Microsoft's analysis of price differences is much different.

Amazon has been more aggressive about dividing both its technology and prices into smaller pieces to appeal to a wider variety of customers, Hackett says. Demand from users has apparently not grown intensive enough for Microsoft to follow suit, but it almost certainly will as both PaaS and IaaS become more widely adopted, he says.

Most recently, Amazon announced a new offering to give customers a free taste of EC2. As of November 1, customers can get one year of free access to a micro-instance on Amazon's Elastic Compute Cloud, plus a limited amount of storage and data transfers. This free ticket to the Amazon cloud can be used for enterprise apps or for test and dev purposes.

Follow everything from CIO.com on Twitter @CIOonline.

2. Application support

While both Azure and EC2 heavily market applications that ISVs make available through their services, EC2 tends to attract resource-intensive software typically classified as enterprise applications, while Azure tends to feature applications that typically serve LAN- or workgroup-sized user groups.

Both platforms are designed to be easily resizable and are being developed quickly; for now at least, that minimizes but doesn't eliminate the difference between IaaS and PaaS, Golden says.

3. Application portability

Because it's a Microsoft environment and supports commonly used x86-based computing technologies such as .NET and Visual Studio, Azure is more accessible to developers accustomed to those environments, according to Gary Chen, IDC server virtualization analyst.

The benefit and drawback to that is that Azure only offers Microsoft operating systems, which makes things simpler for Windows developers, but limits the applications they can build on or port to Azure, Hackett says.

The new Windows Server capability is designed to make applications even more portable between data center and cloud, but only when both are closely equivalent Windows Server environments.

Amazon offers Linux, Mac OS X or Windows support, but has some specific storage requirements and also requires developers to take into account the amount of bandwidth and computing time that the application will require.

4. Scalability

Both Azure and EC2 are designed to be scalable, but the way they achieve this is different.

On Azure users have the choice of scaling up the number of VMs on which an application runs, or by increasing the power of the virtual machine they use, according to .NET development consultant Jason Haley. It offers storage as varying blocks of Binary Large Object (BLOB) service and virtual NTFS drives.

Pay-as-you-go on Azure means defining the number of VM nodes that an application requires and the ability to change that number using a management console.

EC2 is far more flexible now than two or three years ago, Golden says. It still scales in blocks called Elastic Compute Units (ECU) that include a specific number of Amazon Machine Images (AMI) and storage volumes called Elastic Block Stores (EBS).

ECUs include varying amounts of storage, I/O, memory, performance equating to specific 32- or 64-bit processors, and a variety of networking and higher-level systems or application-management services.

At one time Amazon offered only small, medium and large-sized resource blocks and price points. Then in September, it introduced micro instances that require as little as 613MB of memory, and cost as little as two cents per hour of use. An Amazon feature called autoscaling automates changes in resource levels to respond to demand.

5. Price

Because Amazon's service is designed to be more comprehensive, its pricing is more complex than Microsoft's, though neither pricing plan is simple.

Amazon's service is reputed to cost more in general, but developers and analysts also complain about the rigidity and surprisingly high cost of Microsoft Azure, often finding EC2 cost-effective even for small projects when comparing the two. Microsoft's analysis of price differences is much different.

Amazon has been more aggressive about dividing both its technology and prices into smaller pieces to appeal to a wider variety of customers, Hackett says. Demand from users has apparently not grown intensive enough for Microsoft to follow suit, but it almost certainly will as both PaaS and IaaS become more widely adopted, he says.

Most recently, Amazon announced a new offering to give customers a free taste of EC2. As of November 1, customers can get one year of free access to a micro-instance on Amazon's Elastic Compute Cloud, plus a limited amount of storage and data transfers. This free ticket to the Amazon cloud can be used for enterprise apps or for test and dev purposes.

Follow everything from CIO.com on Twitter @CIOonline.


Follow Us

Join the newsletter!

Error: Please check your email address.

Tags amazonMicrosoft

Featured

Slideshows

Channel celebrates as HP marks 50 years in NZ

Channel celebrates as HP marks 50 years in NZ

HP marked 50 years in New Zealand at an event in the vendor's Auckland's headquarters last night, with a host of key channel figures coming along to celebrate. Photos by HP.

Channel celebrates as HP marks 50 years in NZ
EDGE 2017 - Icebreaker Sessions round 2

EDGE 2017 - Icebreaker Sessions round 2

EDGE guests experience the value of networking at the second round of Icebreaker sessions.. Photos by Maria Stefina

EDGE 2017 - Icebreaker Sessions round 2
Show Comments