Managing your data in a hybrid cloud environment
By Mark Bentkower August 27, 2015
- By end-2017, nearly half of large enterprises will have hybrid cloud deployments
- Public and private cloud infrastructures, legacy systems must be fully integrated
AS organisations attempt to strike a balance between having adequate computational services on premises, and offloading extraneous storage and computational capabilities to the cloud, a logical middle ground has emerged: The hybrid cloud.
Many analysts believe that 2016 will be a defining year for the cloud as private clouds make way to the hybrid cloud. And by the end of 2017, it is estimated that nearly half of large enterprises will have hybrid cloud deployments.
In order for hybrid clouds to be effective, the public and private cloud infrastructures, legacy systems, as well as data and workflows, must be fully integrated.
And although the management platform for hybrid cloud solutions is a critical piece of this puzzle, it is also often overlooked.
Here we outline some of the key considerations for companies to take note of in order to leverage the hybrid cloud effectively and efficiently.
READ ALSO: Only 5.7% of Malaysian businesses effective with digital transformation: IDC
1) Know your data
Planning a journey to the cloud, whether private, public or both, is daunting for all organisations.
There is the promise of greater business agility and low upfront investment – however, if not handled systematically and driven by insights gleaned from data, it can actually increase cost and complexity.
In Singapore, for instance, more than a third (38%) of businesses blame the trend of keeping data management and analytics strategies at a departmental level as the root cause of data silos.
By starting with insights from their data, organisations can better understand which workloads and applications are most appropriate for a public or private cloud hosting, and deploy a successful hybrid model.
It is also important to identify the organisation's most valuable data and to prioritise storage management resources appropriately.
Developing and implementing the right data retention policy is a necessity for both internal data governance and legal compliance. Some data needs to be retained for many years, whereas there may be data that are crucial or required for just a few days, and yet others which should not be retained at all.
Data classification is the first critical step to placing the right data in the appropriate storage tier.
Keeping this in mind, organisations can then effectively determine which route to take – be it building on-premises solutions, moving fully to the cloud, or opting for a combination of the two.
2) Get the most out of virtualisation
Virtualisation is a game changer for organisations adopting cloud as a computing platform and/ or as a storage environment. The benefits of server virtualisation are compelling from cost savings due to business flexibility and the agility inherent in emergent private and public cloud architectures.
As IT departments mature in their use of virtualisation, they are bringing increasing numbers of critical applications online.
Yet, they often forget to assess whether their existing data protection platform will be equally effective for those applications as their underlying infrastructure changes.
At the same time, the ease of deploying new VMs (virtual machines) can also lead to a virtual machine sprawl, making it tedious and time-consuming for administrators to keep track of new VMs and to ensure data protection and retention policies are applied to them.
A key priority for organisations should be to ensure their infrastructure can handle the scale, integration, recovery and flexibility that virtualised applications require.
Although traditional backup methods may be fine in many cases, it is essential to remember that every application has its own RTO and RPO (recovery time/ point objective) requirements.
Making sure that appropriate backup methods align with an application's needs can dramatically enhance a virtualisation strategy – meeting even the strictest service-level agreements.
3) A disaster recovery strategy for the hybrid environment
The public cloud is quickly becoming a real option for IT as another class of infrastructure. One place where many organisations are considering public cloud infrastructure is for business continuance and disaster recovery.
That said, many companies still aren’t able to allocate budget to disaster recovery across the existing data centre, and instead, are adopting a siloed approach, budgeting for mission-critical applications only.
Given that the cloud is a compelling option for non-critical functions, a situation is developing in which a significant volume of data could be excluded from any disaster recovery coverage in the cloud. This is typically when the disaster recovery strategy is not applied throughout a hybrid environment.
By leveraging an effective combination of hybrid and public clouds, businesses can not only enhance data availability for critical apps, but also implement broader disaster recovery in more efficient and cost-effective ways.
The new normal
The hybrid environment is fast emerging as the norm for many CIOs (chief information officers).
The challenge now is to integrate and govern such a system, preferably without altering the existing on-premises infrastructure or the applications.
This, in turn, means organisations need common data and software management tools to define where various workloads and applications are to be hosted.
Ultimately, this will enable organisations to reduce costs, improve utilization, and simplify traditionally complex and siloed management to unlock the true potential of cloud projects.
Mark Bentkower is director of Systems Engineering Asean, CommVault Systems.
Best practices for hybrid clouds
Moving to the hybrid cloud in three steps
Darwinism, survival of the fittest and the cloud
Smart data management in the ‘third platform’ era
For more technology news and the latest updates, follow us on Twitter, LinkedIn or Like us on Facebook.