[How to] Be Confident Storing Information in the Cloud
March 29, 2011 | 2 Comments
By Anil Chakravarthy and Deepak Mohan
Over the past few years, information explosion has inhibited organizations’ ability to effectively secure, manage and recover data. This complexity is only increasing as organizations try to manage the data growth by moving it to the cloud. It’s clear that storage administrators must regain control of information to reduce costs and recovery times while complying with regulatory compliance standards, including privacy laws.
Data growth is currently one of the biggest challenges for organizations. In a recent survey by Gartner, 47 percent of respondents ranked data growth as the biggest data center hardware infrastructure challenge for large enterprises. In fact, IDC says that enterprise storage will grow an average of 60 percent annually.
As a result, companies are turning to the cloud to help them alleviate some of the pains caused by these issues.
The Hype of the Cloud: Public, Private and/or Hybrid?
There is so much hype associated with cloud computing. Companies often struggle with defining the potential benefits of the cloud to their executives, and which model to recommend. In short, the cloud is a computing environment that can deliver on-demand resources in a scalable, elastic manner and is typically, but not necessarily, accessible from anywhere – through the internet (https://blog.cloudsecurityalliance.org/). The cloud encompasses the principle that users should have the ability to access data when, where and how they want – regardless of device.
The public cloud is typically when a third party owns the infrastructure and operations and is delivering a service to multiple private entities (i.e., cloud-based email or document sharing). While these services typically provide low-cost storage, this model has a few drawbacks: companies have limited control over implementation, security, privacy. This can be less than ideal for some organizations.
We believe most enterprises will implement a private cloud over the next few years. A private cloud retains control by enabling the company to host data and applications on their own infrastructure and maintain their own operations. This gives them maximum control, protecting against unforeseen issues. Private clouds can be scalable and elastic (similar to public clouds), providing them the best online storage operations and options to improve performance, capacity, and availability as needed.
A hybrid approach enables organization to combine the inexpensive nature of public cloud and private clouds, but giving additional control over the management, performance, privacy and accessibility of the cloud for their organization. For example, an organization may define a private cloud storage infrastructure for a set of applications and take advantage of public cloud storage for off-site copies and/or longer-term retention. This gives the organization the flexibility to deliver a service-oriented model to their internal customers.
Deciding which model to use is crucial. Each organization ought to evaluate their application portfolio, determine the corporate risk tolerance, and look to an agile way to consume cloud services. For small and medium-sized enterprises the propensity for public cloud applications and infrastructure can be much greater than large enterprise organizations.
The Private Cloud and Virtualization: Tools to Minimize Data Growth
As companies look to private clouds, often leveraging server virtualization, to more efficiently deliver applications to their business, it can also help manage, backup, archive and discover information. Adding to this issue, IDC reports that companies often waste up to 65 percent of storage capacity as disk utilization rates range from 28-35 percent. Cloud initiatives seem like the natural solution.
The private cloud is the clear answer. Combined with virtual environments, if managed correctly, the cloud can help organization save money, increase application delivery times, increase application availability, and reduce risks.
As a best practice, organizations need to increase storage utilization within virtual infrastructures as virtual machines deployments can often result in unused and wasted storage. Moreover, there can be performance implications as companies look to virtual desktops when a large number of users log into their desktops simultaneously, performance can suffer dramatically. Organizations can use new tools that address the storage challenges of virtual environments and integrate with the virtual machine management console for rapid provisioning of servers and virtual desktops. This would include the cloning and set up of virtual machines, desktops, applications, and files needed across all virtual servers.
By having intelligent storage management tools, organizations can reduce the number of copies stored for virtual machines and desktops yet still deliver the same number of applications and desktops to the business. This enables administrators to utilize the appropriate storage (including the appropriate characteristics cost, performance, availability, etc). According to our own tests, this can eliminate as much as 70 percent of storage requirements and costs by storing only the differences between VM boot images.
In addition, by utilizing appropriate management tools that look across all environments – whether physical, virtual or cloud-based – organizations can drive down costs by giving them a better understanding of how they are using storage to improve utilization and make better purchasing decisions. Furthermore, using such centralized management tools will help them to better automate tasks to improve services and reduce errors. This automation helps organizations deliver storage as a service (a key tenant for private cloud computing) with capabilities including on-host storage provisioning, policy-driven storage tiering and chargeback reporting.
Another example is when organizations back up applications within their virtual environment, they have normally done two separate backups: one for the full image recovery and one of the individual files within the environment for recovery later. Organizations can reduce this waste by implementing solutions that will do a single backup that is off-host, in the cloud, and will allow them to do two separate recoveries of the full image and of granular files. This more effective implementation of deduplication keeps data volumes lower and allows for better storage utilization.
Hybrid Cloud Solutions: Control of Storage Utilization and Archiving
Data protection and archiving environments within virtual and cloud environments tend to grow faster than anticipated. These environments will need to be managed closely to keep costs down. Luckily, there are software tools that address this quite effectively.
Implementing a hybrid model allows organizations to get storage offsite (through public cloud storage), eliminating tape rotations and other expenses associated with off-premise storage. However, organizations should be cautious when looking at tools that don’t provide consistent management across physic, virtual, and cloud-based infrastructures.
Many organizations are examining cloud-based email such as Google’s Gmail and Microsoft Office 365. But, as a best practice for this hybrid model, organizations can’t compromise corporate security and governance policies. This often results in organizations needing to maintain on-premise email archiving and discovery capabilities with information that resides in the cloud. In doing so, organizations now have a consistent way to discovery information that in their private cloud as well as information hosted in the public cloud.
Of course, organizations that integrate tightly with major cloud storage partners will see the biggest benefit of this hybrid approach – especially if they need to quickly deploy a cloud implementation to meet rapid growth.
Moving Forward with an Eye to the Sky
IDC reports that 62 percent of respondents to a recent survey say that they will be investing in data archiving or retirement in 2011 to address the challenges associated with data growth. IT organizations are in the process of trying to re-architect their environments to meet these challenges. Private and hybrid clouds, combined with virtualization, seem key in addressing these challenges.
By implementing cloud solutions, storage administrators are regaining control of information, helping them to reduce storage costs, and better deal with tomorrow’s challenges.
Anil Chakravarthy, Senior Vice President, Storage and Availability Management Group and Deepak Mohan, Senior Vice President, Information Management Group, Symantec Corporation