As business look at cloud adoption, the question is, “What is the cloud good for?” Yes, the cloud can be efficient and elastic, but what would be its real use in complex environments?
A different way of looking at the road to the cloud is by considering where your data—both primary and secondary copies—resides.
Cloud as 3rd copy
In a typical data center your primary file and application data would reside in on-premises storage arrays such as XtremIO, VMAX, VNX or Isilon. Second-level protection is offered via data protection solutions using secondary storage such as EMC Data Domain. As a last step, select data can be tiered to the cloud via products such as CloudBoost or CloudArray, to either private clouds built using ECS or public clouds. That means that data in the cloud is a 3rd tier of data. This is a good way to achieve efficiencies for specific use cases such as long-term retention, offsite copy of data, and data archiving while maintaining all primary processing within the data center.
Cloud as 2nd copy
A more direct way of using the cloud is by having data copied directly from primary storage to the cloud (that is, storage tiering) or protecting data directly to the cloud. This results in even higher efficiencies; however, this creates a much larger dependency on the cloud for operational recovery purposes as there is no second copy of data on premises.
Ideally, at this point we would look at direct-to-cloud tiering and protection with ability to maintain on-premises copies of active data for quick access.
Cloud as 1st copy
The last step in cloud adoption is where your primary data resides directly in the cloud, either with SaaS applications (such as Office 365, Salesforce, and Google Apps) or hosted applications running on cloud-based PaaS (such as developed using Pivotal’s Cloud Foundry platform) or IaaS (such as Amazon AWS or Microsoft Azure). In that case, the need for data protection still remains even if infrastructure resiliency responsibility has moved to the cloud solution provider. And to achieve efficiencies, data protection solution and resulting data copies also reside in the cloud itself. However, in that case you may need to export data back to on-premises either for safety, compliancy, or other reasons.
Which is the right approach?There is no single answer as it primarily depends on the application. Although applications that have perfect-fit and ready-made SaaS offerings are good reasons for adopting cloud versions instead of managing private servers, traditional applications that reside in the data center may stay there. But some of those traditional applications could also be moved to cloud-based IaaS as well. And new 3rd platform applications can be developed using new PaaS solutions and be more agile in their operational model, either on-premises or in the cloud.
So what’s the answer? It depends.
As you consider which approach is best for your organization, be aware that EMC provides all the solutions necessary for a multi-faceted IT environment:
- Tiering to the cloud for both storage arrays as well as data protection solutions via products and technologies such as CloudBoost, CloudArray, Cloud Tiering Appliance, Isilon, CloudPools, and soon additional solutions as well.
- Data protection to the cloud via products such as MozyPro and MozyEnterprise.
- Data protection in the cloud for SaaS applications via products such as Spanning for Office 365, Salesforce and Google Apps.
- Data protection in the cloud for traditional applications using NetWorker and CloudBoost and soon additional solutions such as native cloud copy management
What about the future of cloud services? Although no one can predict the future with perfect accuracy, you can be sure that EMC will be on that road to efficiency!