Guide To IaaS Cloud Provider Performance
Choosing an Infrastructure-as-a-service cloud provider? Use these benchmarks to compare AWS, Google, Azure, Rackspace, and more.
July 23, 2014
![Network Computing logo Network Computing logo](https://eu-images.contentstack.com/v3/assets/bltde8121fc52c5c8f3/blt3f3d0318f746b1c2/65a530e4187606040a1d8b8c/placeholder.png?width=1280&auto=webp&quality=95&format=jpg&disable=upscale)
I've been frustrated by a lack of comprehensive comparison benchmarks between Infrastructure-as-a-Service (IaaS) providers and within providers themselves. Most performance data shows only a small number of available "instance types" in a limited region, with no way to compare pricing or application performance.
As an Amazon Web Services customer, I know from my own experience that the c1.xlarge performs really well for many applications, but no one appears to have ever compared that to other cloud services.
So, I decided to run my own IaaS performance project and benchmarked every instance type across the major public IaaS provider regions on a consistent setup. My primary focus was to provide a broad, comprehensive view of most serious compute options available to IaaS customers.
Part 1: Test Methodology
First, I explained my goals for the benchmarks and how I used the tools UnixBench and SysBench’s MySQL to test different instance types in different data centers across different providers.
Part 2: Amazon Web Services
I launched more than 175 AWS VMs across instance types and compared performance by region and availability zone. Here you can read the detailed results.
Part 3: Amazon Web Services C3
After Amazon rolled out the C3 family of instances, I ran the benchmarking tests using C3 and updated my AWS charts accordingly.
Part 4: Google Compute Engine
I compared instance types across Google Compute Engine and against AWS, finding it a solid competitor to Amazon's offerings.
Part 5: Microsoft Azure Virtual Machines
I tested Windows Azure Virtual Machines and compared Azure to AWS and Google Compute Engine. Here you'll find the results.
Part 6: Rackspace Cloud Servers
I evaluated Rackspace Cloud Servers' compute resources and how they compare to AWS, Google Compute Engine, and Azure. See how they measure up.
Part 7: SoftLayer
In this part, I looked at SoftLayer, an IBM company, which offers bare-metal (not virtualized) pay-by-the-hour services.
Part 8: HP Public Cloud
I tested HP Public Cloud across regions and compared it to the other IaaS providers I've benchmarked. HP has recently focused on providing an OpenStack-based public cloud.
Do you have experience with these IaaS providers? Let us know how they fared in the comments.
About the Author
You May Also Like
Maximizing cloud potential: Building and operating an effective Cloud Center of Excellence (CCoE)
September 10, 2024Radical Automation of ITSM
September 19, 2024Maximizing Manufacturing Efficiency with Real-Time Production Monitoring
September 25, 2024