How Software Defined Storage Helps Cloud Computing Service Providers Overcome Today’s Challenges
Every organization that depends on its IT structure has thought about the cloud at some point in 2016. Cloud computing is a naissance that is fueling the digital transformation of the world and the way it does business. The term cloud is an abstraction for the complex infrastructure it conceals. The generally accepted definition of cloud computing comes from the National Institute of Standards and Technology (NIST). The NIST definition runs to several hundred words but essentially says that:
”Cloud Computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources that can be rapidly provisioned and released with minimal management effort or service provider interaction.”
Cloud computing is about agility and flexibility such as the ability to meet workflow demands in real time by increasing and decreasing capacity as needed in automated, usually API & policy driven fashion. By migrating resources and services off premise, companies can better respond to both opportunities and crisis in real time. This new paradigm approach to computing accommodates real reductions in CapEx spending by moving from a fixed cost model to a variable cost one. In doing so, organizations only pay for the resources they use, allowing them to divert investment capital to revenue generating projects.
Unless of course, you are a service provider providing cloud computing services for your customers. As a service provider, you must provide the IT infrastructure and face the very same challenges that your customers are turning to you in order to avoid.
Consider these challenges:
- You have to risk a great amount of fixed capital expenditure in order to provide variable cost model benefits to your customers.
- In order to provide scalability to your customers, you must host excess infrastructure resources that sit idle, a complete contradiction to the very premise of cloud computing.
- You must constantly fight technology obsolescence in order to deliver the benefit of ensuring delivery of the latest technologies to your customers.
- You must provide a fertile ecosystem that will allow your customers the means to automatically deploy their own services and resources.
- You must provide the necessary resources that will meet the stringent expectations that users have today when it comes to issues of performance.
- Security is a pinnacle concern when hosting customer resources in a multi-tenant environment.
- And you must do this while competing with larger competitors who enjoy economies of scale that are unobtainable for you.
The challenges can be daunting.
Obviously cloud computing depends upon server virtualization which maximizes compute resources such as memory and CPU and consolidates server fleets. Because of fast advance of hardware and improvements in virtualization, host servers today can easily accommodate tens to hundreds of virtual machines (virtual servers), obviously depending on the size of these virtual servers. Now virtualization and cloud management software allows new virtual machines to install in a matter of seconds and can boot nearly instantaneously. We are also finally seeing the fruition of software defined networking solutions today that allow true separation between the control plane and data plane, allowing for automated network device deployments that are then governed and managed by preconfigured delivered policies.
But the biggest improvements lie in the area of storage. It is the most important component of your infrastructure as it hosts the data that drives innovation and knowledge for your customers. Also, in most cases, it is the bottleneck of the cloud solution and solving it impacts the whole cloud. And reliability is top of mind in the multi-tenant environment of a service provider, which demands shared storage. Although shared storage has been around for more than a decade, it has been characterized as a proprietary monolithic hardware centric structure, unsuitable for the elastic fluid nature of cloud computing. The solution – Software Defined Storage.
Here is how SDS overcomes the challenges that cloud service providers must overcome:
- Capital Expenditures (CapEx) – Software Defined Storage is built around x86 storage technology which is far less expensive than the proprietary storage arrays found in the SAN solutions of yesteryear. By utilizing a universal hardware standard, service providers can lower acquisition costs as well as TCO.
- Utilization – a really fast storage solution (latency is key) can increase server utilization 3 to 4 times, meaning a cloud service provider will be able to host 3-4 times more VMs on the same underlying physical hardware (usually with some minimal upgrades, typically RAM only).
- Flexibility – Unlike a SAN solution, an SDS can be composed of heterogeneous hardware that can constantly be modified in order to meet changing demands. This means that high value critical data is matched with high performance flash drives while less significant data can reside on less expensive media.
- Scalability – The reduced cost structure of x86 storage technology coupled with the flexibility of multiple performance offering allows service providers the ability to rapidly update and expand their storage structures to accommodate future growth in fast turnaround
- SaaS – SDS totally redefines storage as a service, fully negating the constraints and defined limitations of storage hardware. Instead, SDS delivers storage with the agility that only software services can achieve.
- True Centralized Intelligence – Like all software defined solutions, SDS separates the intelligence from the physical devices. This greatly simplifies administration and allocates efficient and secure management of all storage silos through automated policies that can accommodate and react to ever changing conditions.
All of these benefits reduce the time to value of your infrastructure investment. SDS can accelerate and deliver a highly competitive ROI advantage for your organization. It’s for these advantages and proven financial returns that Gartner estimates that nearly 30% of the global storage array capacity installed in enterprise data centers will be deployed with integrated storage system architectures based on x86 hardware by 2019.
If you have any questions feel free to contact us at [email protected]
Share this Post