Introduction
We recently began benchmarking Lunacloud, an IaaS vendor based in Europe. We have found strong disk IO and networking performance for Lunacloud. Amazon’s performance remains unpredictable with periods of spikes, while Rackspace Cloud (not OpenStack Cloud) performs in a stable manner throughout the 30 days.
Findings
We tested on the following server configuration: 4 vCPUs, 16 GB RAM, 50 GB disk. While Lunacloud outperforms Amazon and Rackspace in many cases over the 30-day test, the service also prices itself much lower for such a configuration.
Key Findings
- In the general server test, Lunacloud outperforms Amazon and Rackspace consistently over a period of 30 days.
- Lunacloud disk performance scores up to 8x better than Amazon or Rackspace disk performance. This is an important consideration for database performance.
- Lunacloud’s internal network throughput is 2x faster than Rackspace’s internal network throughput, and significantly more stable than Amazon in a period of 30 days. With applications requiring more than one server, internal network speed may be considered a bottleneck in many cases if other components of the server are optimized.
- Amazon’s EC2 CPU performance generally scores slightly better than Rackspace and Lunacloud CPU performance. CPU is a vital component of the server, and application performance depends significantly on CPU functionality.
- Rackspace’s RAM is more consistent in performance than Amazon or Lunacloud, though it scores the lowest in performance over a period of 30 days. Though the processor speed is an important factor in server performance, RAM plays an equally important role. Because modern CPUs can perform many of the applications run today, bottleneck instances may actually occur with RAM.
Snapshot of Findings
To enlarge an image, please click on it below. To download a full report on our analysis, please click here.
General Server Comparison using UnixBench:
Amazon Average 30-day Score: 1,037 points
Rackspace Average 30-day Score: 955 points
Lunacloud Average 30-day Score: 1,305 points
Amazon CloudSpecs Score: 40
Rackspace CloudSpecs Score: 26
Lunacloud CloudSpecs Score: 100
The purpose of UnixBench is to provide a basic indicator of the performance of a Unix-like system; hence, multiple tests are used to test various aspects of the system’s performance. These test results are then compared to the scores from a baseline system to produce an index value, which is generally easier to handle than the raw scores. The entire set of index values is then combined to make an overall index for the system.
Methodology
As mentioned earlier, we monitor performance 4 times a day, 365 times a year with our CloudSpecs system, a software suite of open-source, industry-standard server performance tests. From those findings, we average performance in a period of time (for purposes of this post, it is 30 days). Using that average, we plug in pricing to figure out a value score based on how much the server cost us and how much performance we’re actually getting out of it. The best value IaaS cloud provider is given a score of 100, and every other IaaS provider has a score pegged to it, so the other IaaS providers’ scores are in relation to the best IaaS provider’s value.
Price:Performance CloudSpecs Score calculation methodology:
- provider_value(P) = [Provider test score over a period of time] / [Provider price]
- best_provider_value = max(provider_values)
- Provider’s CloudSpecs Score = 100 * provider_value(P) / best_provider_value
Pricing/Server Cost:
- Amazon’s XLarge Instance: $0.64 per hour
- Rackspace’s 15GB Server: $0.96 per hour
- Lunacloud’s Server: $0.32 per hour
Server configuration is a bit tricky because both Amazon and Rackspace only provide tiered structures, where you must select a pre-configured package (called “instances” on Amazon and “servers” on Rackspace) that contains a set amount of CPU, RAM, and disk space.
Server Configurations:
- Amazon: 15GB RAM, 4vCPUs (8ECUs), 1,690GB Disk
- Rackspace: 15GB RAM, 4vCPUs, 620GB Disk
- Lunacloud: 16GB RAM, 4vCPUs, 50GB Disk
For more details into the methodology, please contact us.