The 7th Tenet of Hybrid Cloud: An Open Hybrid Cloud Requires Open Hybrid Storage

The Biggest Hybrid Cloud Challenge?

A while back, I met with a customer in the media industry that was tackling the problem of how to make movies in a worldwide hybrid cloud. This company had multiple datacenters around the world and also leveraged the public cloud for spill-over capacity to render its movies and effects. Furthermore, because this company worked on several films at a time, it had a huge volume of media assets it needed to replicate around the world—images, sound, movie frames, etc. This company could quickly spin up compute capacity in all its different datacenters that was configured identically. But, how could it replicate its data for those compute jobs around the world in a hybrid cloud?

This problem is not unique to the media industry: every enterprise that wants to leverage the public cloud must wrestle with the challenge of how to get its data into the public cloud. And, they must not only get their data into the public cloud, they need to keep that data consistent for workloads in their private datacenters and clouds.

This, then, is perhaps the biggest challenge for enterprises that want to adopt the cloud—how do you access your data across a hybrid cloud environment? If an enterprise can spin up identically configured virtual machines in a public cloud and in its own datacenters, but all of that company’s data is located just in its own datacenters, or just in the public cloud, that ability to launch identical virtual machines across a hybrid cloud is useless because the workloads won’t run across these various locations. Unless you can share your data uniformly across your cloud, you are going to create a bunch of small cloud islands, each with a large data access moat separating it from the rest of your environment.

This leads us, then, to the 7th tenet of hybrid cloud: an open hybrid cloud requires open hybrid storage.

Red Hat Storage Redefines Storage For the Open Hybrid Cloud

Red Hat Storage is a software-defined storage product based on the Gluster open source project. It runs as a pure software application on top of Red Hat Enterprise Linux and commodity x86 hardware. This in itself provides significant value over proprietary storage systems that require expensive, specialized hardware. But, the true power of Red Hat Storage becomes evident in a hybrid cloud environment:

  • Scale-Out Storage: Red Hat Storage provides a horizontally scalable storage solution for the cloud. Red Hat Storage shares and replicates data across multiple systems. The more systems that you add, the more storage capacity you add to your storage pool. Now, in a cloud environment, adding systems is a trivial task—you just launch new virtual machines. And, because Red Hat Storage can run in virtual machines, scaling out storage capacity is a flexible and dynamic task with Red Hat Storage in the cloud.
  • Public Cloud Ready: Because Red Hat Storage is a pure software application, it can run easily in the public cloud. For example, Red Hat Storage is available at Amazon AWS, and soon enterprise customers will even be able to take a free Test Drive of it there. Try doing that with a hardware-based storage solution! You can’t, because you need a software-defined storage system to bring your data to the public cloud.
  • Global Namespace: Red Hat Storage not only provides the ability to scale out data across systems within one rack or datacenter, but it can do this across geographies and public/private cloud boundaries—all with a global namespace. And, Red Hat Storage includes fast geo-replication so that data changes made in one location propagate quickly to another location. This means that if you build a hybrid cloud with, for example, private cloud datacenters in New York and Los Angeles, and a public cloud at Amazon, you can launch virtual machines in any one of those locations, and they will all have access to and be able to manipulate the same data in the same global namespace. It doesn’t matter where the workloads run, they will have a consistent view of the world.
  • Multiple Ways To Access Data: Red Hat Storage not only bridges different locations for your data, but it also spans multiple ways to access your data. Today, most applications expect to access data through traditional storage interfaces such as POSIX. Red Hat Storage provides a POSIX interface, so you can mount it as a standard filesystem over NFS or CIFS. But, many new cloud applications today are also leveraging a new paradigm of object-store interfaces. Red Hat Storage supports that too by offering native support for OpenStack Object Storage (Swift) (in fact, Red Hat Storage has many integration points with OpenStack). And, Red Hat Storage provides the ability to access the same data simultaneously across these different client types. This means that you can bring existing applications and new applications into the cloud—and provide them all with access to the same data.
  • Big Data Optimized: One of the most prominent workloads in cloud today is big data. Big data is ideally suited for an open hybrid cloud because it requires scale-out storage, compute, and applications—all of which the cloud can deliver well. Red Hat has been integrating Red Hat Storage with OpenStack and Hadoop to build a powerful cloud-based foundation for big data applications

Application Portability Is Key

An open hybrid cloud must provide application portability across heterogenous, hybrid environments. Red Hat provides a comprehensive set of capabilities addressing interoperability and portability across a hybrid cloud:

  • Red Hat Enterprise Linux provides a portable application runtime environment across various clouds
  • Red Hat JBoss Middleware provides a portable application container and framework across various clouds
  • Red Hat OpenShift is a PaaS that runs both as an online service and in private data centers, providing a portable PaaS across various clouds
  • Red Hat CloudForms provides a hybrid cloud management platform with unified self-service and operational management across various clouds

Together, these offerings provide hybrid cloud portability and interoperability from virtual machine setup through application environments to the management layer. Red Hat Storage completes this application portability picture by providing data portability across a hybrid cloud. Combined, Red Hat is able to deliver an open hybrid cloud with true application portability across diverse, geographically separated environments.

About bryanche

I am responsible for overall cloud and product strategy at Red Hat.

Trackbacks/Pingbacks

  1. The 4th Tenet of OpenStack: Open Source Projects Are Not The Same as Products | tentenet.net - November 21, 2013

    […] Hat Storage, which provides an open software-defined storage solution for clouds and plugs into […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 872 other followers

%d bloggers like this: