Rackspace catches up to Amazon... four years later: Say hi again to auto-scale
New demand-response feature is vintage Bezos tech
Rackspace has announced the general availability of "Auto Scaling" technology for its public cloud, catching up with Amazon which implemented the tech in 2009.
Auto Scaling is a crucial component of any cloud, as it allows new resources to be allocated to live virtualized servers in real-time in response to demand, and without the need for a shutdown.
Rackspace's version has both schedule-based and event-based policies, so admins can either define time periods during which they want infrastructure to scale up, or use monitoring services to create auto-trigger events.
"Auto Scale is the easiest way to make your Rackspace Cloud react automatically to changes in the demand your users place on your application," wrote Rackspace employee J R Arredondo.
By implementing the technology, Rackspace has added another valuable feature to its public infrastructure-as-a-service cloud, but it also highlights the gulf in capability between the Texan web-hoster-turned-cloud-hustler, and Amazon Web Services, which launched the scaling tech in May 2009.
Rackspace can take solace in the fact that other major rivals, such as Microsoft via Windows Azure and Google via Compute Engine, have also been slow to implement the much-needed feature.
Microsoft announced built-in Auto Scaling in June. Google, meanwhile, has made the tech possible, but it's difficult to create and requires admins to use a broad swath of the Chocolate Factory's arcane infrastructure.
Since Amazon made Auto Scaling available on its sprawling platform several years ago, we do find ourselves wondering why it took these companies so long to implement. One possible reason is that the heavily modified Xen hypervisor used by AWS could be easier to stretch than the systems in use at Google (KVM) and Microsoft (Hyper-V). But that still doesn't explain how Rackspace (Xen), has been slow to implement the tech. ®