OpenStack can impact just about every aspect of an IT infrastructure, including storage.
Storage continues to gain importance commensurate to the rise of enterprise data. Big data, with information measured in petabytes, has become a reality for many organisations, and workers must be able to access information quickly via multiple devices.
This requires a storage infrastructure with an enormous amount of flexibility and agility – the same attributes that have made OpenStack popular.
That approach is in direct contrast to what traditional storage infrastructures offer. Historically, enterprises have relied on large, monolithic, hardware-based storage appliances.
Maintenance of these storage “mainframes” involves tedious management and painful upgrades when they inevitably become maxed out. They are the anti-agility solution.
Given enterprises’ current storage needs, it makes no sense for CIOs to pair OpenStack with monolithic hardware appliances. This approach is a recipe for frustration and ineffectiveness because it risks undermining everything these executives have set out to achieve with OpenStack.
A whale vs. a school of fish
Let’s be blunt: there’s simply no place for large and bloated storage systems in today’s enterprise, which strives for agility and scalability.
In fact, comparing traditional storage architectures to today’s open and software-defined options is like comparing a whale to a school of fish. Both can ultimately do the same thing with the same amount of power, but one is far more fluid and adaptable than the other.
Fluidity, adaptability and power are what today’s enterprises are seeking. CIOs want to be able to easily move, process and store data. To do this, they are moving to standardised models that enable their organisations to become more nimble.
It makes no sense for organisations building their cloud infrastructures on OpenStack to choose a storage solution that isn’t flexible. Many of these companies have moved beyond Gartner’s Mode 1 model, which favours traditional and sequential IT practices, to Mode 2, which emphasises agility and speed.
Putting storage management into the hands of people who have not fully embraced the flexibility of cloud architectures is antithetic to their business practices.
Mode 2 leaders want to work with cohesive teams of people with deep cloud experience and the ability to be agile. They do not want to trust their storage to those who take traditional, linear approaches or are encumbered by outdated processes.
In short, using traditional storage is not conducive to increasing agility and cost-effectiveness. It’s simply a bad fit that can create a storage infrastructure that is unable to scale with rising data demands.
An open relationship
Many OpenStack users have found that the next generation Ceph software-defined storage platform can complement their cloud deployments. Ceph is a distributed object store and file system that has been adopted by 62% of OpenStack users.
This is not terribly surprising when one considers the close relationship between OpenStack and Ceph. The two offerings can work together effectively to accomplish data centre virtualisation and business acceleration.
Specifically, OpenStack and Ceph can work together to help enterprises achieve better scalability, economy and flexibility. These are all things which storage “mainframes” are not equipped to deliver.
For example, traditional storage solutions cannot scale to accommodate petabytes of data, whereas Ceph is able to easily handle this amount of information. Its open scalability effectively eliminates the limitations imposed by legacy storage appliances.
Use of Ceph can also be cost-effective, particularly as scale increases. Organisations can use it to create a foundation based on an initially low cost per gigabyte that decreases as more data is used. It’s like buying in bulk – the more gigabytes or petabytes there are, the lower the storage management costs become.
Additionally, both OpenStack and Ceph are open-source solutions, making them flexible and powerful. They can be used in conjunction with any type of hardware and in any kind of environment.
OpenStack and Ceph users also enjoy the support of a vibrant open-source community and broad ecosystem, which makes the customary risk of vendor lock-in a distant memory.
A prime example of this ecosystem is the expanded hardware infrastructure available to Ceph, including flash drives and solid state disks.
Also, a number of partners are already making significant progress in enabling Ceph in their environments, thereby offering greater customer choice and options for deployment based on workload requirements.
There’s a growing demand for OpenStack and next-generation storage solutions. Both represent parallel steps forward in the way CIOs deploy their infrastructures.
Sourced from Sarangan Rangachari, VP and GM, storage and big data, Red Hat