Secondary storage is quickly consuming the data center and becoming too complex to manage. The copy data stored on those secondary systems is distributed across multiple enterprise and cloud storage locations. The secondary data set is being used for a growing number of use cases, and it needs to be organized to adhere to new data privacy regulations as well as eDiscovery requests. In addition, organizations are looking to repurpose secondary data to respond to business needs such as test/dev and analytics.
Recovery expectations are also higher than ever – zero data loss and instantaneous recovery is the standard in many cases. The problem is recovery service level agreements (SLAs) are far from standard; they vary based on application, geography and use case.
To meet the demands on secondary data sets and new use cases, many organizations are looking to the cloud to bypass the capital and operating expenses associated with on-premises data protection. However, many cloud data protection solutions fall short in critical areas such as facilitating recovery and minimizing the amount of on-premises infrastructure required.
Requirements for Enterprise-Grade Data Protection-as-a-Service
To solve the problems associated with an on-premises data protection implementation, enterprises require a solution that tiers older data, which is not needed by the business immediately following an outage, to the cloud. Tiering data instead of mirroring data to the cloud helps to reduce the on-premises infrastructure footprint and avoids paying for additional capacity, both on-premises and in the cloud.
Another important characteristic of an enterprise-grade data protection-as-a-service strategy is the ability to leverage cloud compute in addition to cloud storage. The cloud offers dynamic compute cycles that are ideal for disaster recovery. Disaster recovery-as-a-service (DRaaS) can extend the savings realized by reduction in on-premises infrastructure by also reducing, and potentially eliminating, disaster recovery site costs. However, storage professionals should evaluate DRaaS solutions to ensure they can meet recovery point objective (RPO) and recovery time objective (RTO) SLAs. For example, some cloud disaster recovery solutions first require copying data to a faster storage architecture, which ultimately slows time-to-recovery. Others require downtime when migrating the application back on-premises so that data can be copied back from the cloud.
DRaaS is attractive, but the cloud adds latency that some applications cannot tolerate. To overcome this latency, mission critical applications should maintain a protected copy of data on-premises, as well as in the cloud, for rapid recovery. The problem is most data centers have an increasing number of “important” applications. These applications can’t wait for a cloud recovery but if they are included in the on-premises appliance, it defeats the purpose of an infrastructure-less design. Most data protection applications don’t have a middle option. IT planners should look for the ability to stream virtual machines from the cloud service provider to the on-premises data center, so that applications become available to the user in real-time while data is restored in the background.
Recovery is, of course, the litmus test for any data protection-as-a-service (DPaaS) solution but capturing a useable copy of data is the first step. DPaaS should also include rapid, frequent capture of application-consistent backups, in order to meet stringent RPOs (minimal data loss). In order to meet stringent RTOs, they should also store data in a native format so that no conversion is required on recovery. Storing data copies in their native format also has the additional advantage of avoiding the enterprise being locked in to the data protection provider’s solution until the copies have expired.
Finally, subscription-based pricing is a hallmark of an enterprise-grade DPaaS solution. Data protection costs are increasingly unpredictable, and in an on-premises model, they require the upfront purchase of a large amount of infrastructure. Unfortunately, it then sits underutilized waiting for the amount of backed up data to grow. Each subsequent upgrade brings the same challenge. The customer has to periodically overbuy and tolerate idle resources. The subscription model provides a more on-demand and predictable cost model that helps to optimize budgets.
Conclusion
The software-as-a-service (SaaS) model stands to alleviate not only capex-related headaches associated with data protection, but it can also help to significantly lower the operational overhead – further cutting costs and freeing up IT to focus on more strategic initiatives. At the same time, it can provide more recovery options. To truly better position the storage professional to deliver on the demands of modern RTO and RPO SLAs, however, requires a carefully architected approach that uses the cloud intelligently.
To learn more, watch our on demand webinar “How to Create an Infrastructure-less Backup Strategy”.
Senior Analyst, Krista Macomber produces analyst commentary and contributes to a range of client deliverables including white papers, webinars and videos for Storage Switzerland. She has a decade of experience covering all things storage, data center and cloud infrastructure, including: technology and vendor portfolio developments; customer buying behavior trends; and vendor ecosystems, go-to-market positioning, and business models. Her previous experience includes leading the IT infrastructure practice of analyst firm Technology Business Research, and leading market intelligence initiatives for media company TechTarget.
StorageSwiss.com - The Home of Storage Switzerland 