2015 Tech Trends: A Critical Crossroad for IT
As part of an effort to make updates to operating systems running in cloud computing environments much simpler, Canonical today launched “snappy” Ubuntu Core, a new rendition of Ubuntu that replaces the traditional cumbersome approach to adding updates to an operating system with a simpler transactional model that allows elements of the operating system to be updated in isolation from one another.
Canonical founder Mark Shuttleworth says the benefits of this approach include everything from providing higher levels of security to making it possible for developers to build next-generation applications that take advantage of emerging micro-services based on containers faster.
In concept, the new Ubuntu Core functions similarly to how operating systems on mobile computing devices operate. Updates to those operating systems happen faster and much more frequently than on traditional desktops or servers. Shuttleworth says that Canonical developed its transactional update model as part of an effort to create an instance of Ubuntu for mobile computing devices, and is now applying that model more broadly.
Canonical is not the only provider of an operating system to head down this path. In fact, Red Hat and Google have both created operating systems that embrace a transaction model that not only makes it easier to update operating systems, but also provides the capability to rollback those updates whenever necessary.
Shuttleworth notes that transactional models are the latest in a series of advances that are transforming how operating systems function as a layer of abstraction above the underlying infrastructure. Originally designed to support some level of interaction with users, operating systems in the cloud will increasingly become not only lighter-weight, they will automate more functions.
The end result, says Shuttleworth, will not only be operating systems that consume much less in the way of physical IT infrastructure resources, but also environments that can scale across thousands of servers with the minimal amount of human intervention possible.
