Gartner: AWS, HP have worst cloud SLAs
The award for the "worst SLA" of any major cloud provider goes to Amazon (NASDAQ: AMZN) Web Services, according to Gartner analyst Lydia Leong, but HP's (NYSE: HPQ) public cloud SLA is not necessarily better. Both providers have strict requirements for users in architecting their cloud systems if they want the SLAs to apply when service is disrupted, reports Brandon Butler at Network World.
"Amazon's SLA gives enterprises heartburn," Leong wrote in a recent blog post. "HP had the opportunity to do significantly better here, and hasn't."
In addition to the costly service architecture requirements, the SLAs for AWS and HP cloud services are unnecessarily complex and limited, she complained. For example, neither SLA covers block storage services.
For the AWS SLA to take effect, customers are required to build their systems so that applications run across a minimum of two of the provider's data centers, known as availability zones. The HP SLA only goes into effect if all of its availability zones are down, which means customers have to build their applications to cover three or more availability zones.
The costs of running applications across more than one availability zone add up. What's more, Leong wrote, the requirements add complexity to the systems. "Most people are reasonably familiar with the architectural patterns for two data centers; once you add a third and more, you're further departing from people's comfort zones, and all HP has to do is to decide they want to add another AZ in order to essentially force you to do another bit of storage replication if you want to have an SLA," she wrote.
Because of the service architecture requirements, it isn't likely that AWS or HP customers would be reimbursed for downtime in a meaningful way under the SLAs, Leong warned. However, other infrastructure-as-a-service providers offer considerably better SLAs.
- see Brandon Butler's article at Network World