Filed in Product & Development by Jonathan Hogue | September 10, 2012 3:00 pm
I’ve seen at least a half-a-dozen different definitions for hybrid cloud. If you forced all of the cloud experts into a room and demanded a clear definition, they would only agree on one thing: hybrid cloud is some combination of dedicated compute resources and shared compute resources. The shared resource refers to a public cloud, while the dedicated resource could be on-premise private cloud, hosted servers, off-premise private cloud, etc, etc, etc… as long as it’s reserved for your use. No matter the combination hybrid cloud delivers the flexibility and elasticity of the cloud without having to sacrifice the performance and security of dedicated equipment.
Since there is no completely clear definition for hybrid cloud, the best I can do to explain it is to describe how you can build a hybrid cloud at Rackspace. We have a unique enabler called RackConnect. It’s our organically grown spin on hybrid technology that creates a bridge between your dedicated, hosted network and the Rackspace public cloud. RackConnect lets you store your most sensitive data (e.g. payment card information and customer records), and run your most demanding applications (e.g. large databases or financial applications) on dedicated gear; while you run applications that are well-suited for the cloud (e.g. Database-driven web, messaging and collaboration applications) in the cloud. It brings the two solutions together by allowing them to communicate with each other.
RackConnect has two basic parts: a network device or set of devices, and an automation system. The network device can be a firewall, a load balancer, a combination of the two or a high availability combination (so up to four devices if you’re doing the math). The device(s) passes network traffic between the dedicated and cloud environments in a secure and efficient manner. For example, a firewall can be configured to allow a dedicated database to speak with cloud web servers, but not the Internet.
The second component, the automation system, does most of this configuration for you. When you add new cloud servers to your environment, the system automatically configures the network device to enforce the network security policies you define. When you remove a cloud server, it cleans up. It also configures the software firewall on each of the cloud servers, so that they can only communicate in the way you desire. For example, the software firewall can be set to allow your cloud servers to only communicate with your dedicated servers and the Internet, but not between each other.
To further evolve this solution and continue to meet the needs of our customers, we’ve recently built in some new RackConnect capabilities. Now it can to do even more. RackConnect now has an Application Programming Interface (API) that can be used to retrieve RackConnect status information, such as gateway IP addresses, the automation status, automation status details and the automation features status (whether automation is turned on or off). It also supports Rackspace Cloud Servers powered by OpenStack. New cloud servers can be automatically added to load balancer pools, based on the server name or metadata.
This is just the first of several articles planned for the next few weeks that will cover hybrid cloud. Some of the articles will also announce opportunities for you to come hear us talk about our growth in this space. And if you’re in the Chicago area this week, stop by and see us at Cloud Connect! Also, sign-up to attend our Hybrid Clouds speaking session on Thursday, Sept. 13 at 11:30 a.m.
Source URL: http://www.rackspace.com/blog/building-a-hybrid-cloud-with-rackconnect/
Copyright ©2013 The Official Rackspace Blog unless otherwise noted.