Murrstock - stock.adobe.com
Organizations that need fast, reliable and scalable data protection may find a suitable tool in cloud-native backup.
Cloud-native backups provide comprehensive configuration and data protection for workloads deployed on containers, VMs and hyperscaler services. They are widely supported by cloud vendors, so there are numerous options for organizations of different sizes to choose from.
"When we think of backup and recovery in the cloud, traditional methods need to evolve," said Nicholas Merizzi, a principal at business advisory firm Deloitte Consulting. The ability to preserve data in order to fulfill disaster recovery (DR) needs, meet enterprise and regulatory requirements, and ensure ransomware protection are still very much necessary in the cloud. "Cloud-native backup … leverages native cloud capabilities to achieve the same outcomes in a more operational and cost-effective way," he said.
How it works
"The purpose of cloud-native backup is to allow cloud-native workloads to recover from failed deployments, restarts, failovers and outages," said Mike Williams, managing delivery architect for business consulting firm Capgemini North America. "Additionally, the backups are also used for tracing, auditing or compliance purposes."
The key difference between traditional and cloud-native backup is the ability of cloud-native technology to serve as a more efficient and agile option, Williams said. "[It] enables on-demand access for organizations and developers to make high-impact changes frequently and predictably without worrying about how their data will be stored, protected, recovered and restored," he said.
Cloud-native backup services are consumption driven. "They are supplied by the cloud provider [and] are straightforward to integrate into the environment," said Randy Armknecht, a managing director and global cloud practice leader at global business advisory firm Protiviti.
Benefits of cloud-native backup
Supported by most major cloud providers and many smaller ones, cloud-native backup enables adopters to automate tasks that were traditionally siloed and required specialized skill sets and bulky tools. "This helps in promoting a shared responsibility culture with a 'you build it, you own it' mindset for managing day-one and day-two concerns, while reducing the total cost of ownership and improving speed to market for new product launches," Williams said. "The most important benefit of this approach is that it enables cloud storage and data protection management via code, aligning it with cloud-native paradigms and not sacrificing or exposing data."
Cloud-native backup products and services are frequently self-managed and don't require any form of large operational oversight. "Native offerings carry the benefit that they are consistent, even across multiple cloud regions or different accounts," Merizzi said. The technology also does away with the practice of embedding agents. Access is typically controlled by role-based identity and access management and is an API-driven service. "Through the use of APIs, one can automate or adjust backup policies," he said. "The economics are also beneficial, [with adopters] only getting charged for what is used with no fixed pricing."
Metadata plays a critical role in enhancing the effectiveness of cloud-native data's protection and data governance. Traditionally, a backup application has used metadata to manage and index the massive volumes of data it protected. "While that's still the case today, solutions are now enriching the metadata to enable more advanced capabilities," Merizzi said. "For example, by using time-indexing metadata, recent data can be stored in a faster tier of storage; and, as it ages, it can automatically move to less expensive cloud tiers." Some offerings now supply full-text metadata indexing to speed data discovery, he said.
Use cases and providers
As organizations evolve and become more disciplined around various development aspects -- such as using infrastructure as code and general configuration management -- they are able to shift away from performing tasks such as backing up the root volumes, and instead focus on protecting the data, Merizzi said. "Another interesting use case is being able to easily recover whole instances, applications or a set of containers across multiple regions very quickly," he said. "This cross-region capability, often offered natively in cloud platforms, makes it easier to meet compliance and DR needs with both local and geo redundancy."
Cloud-native backup also provides adopters with nondisruptive data protection. "Cloud-native workloads operate in a continuous fashion, and a data protection solution should operate transparently without a need to disrupt workload operations," Williams said.
Organizations with highly distributed workloads stand to benefit greatly from cloud-native backups. "Cloud-native workloads tend to be highly distributed and dynamic, and a data protection solution needs to operate as a code in conjunction with such workloads," Williams said.
When shopping for a cloud-native backup service, the cloud market leaders are also the top providers, Armknecht said. "If the backup solution is not from a cloud provider, then it's not considered 'native.'"
Mike WilliamsManaging delivery architect, Capgemini North America
Some examples of cloud-native providers include Rubrik, SolarWinds, Veeam and Velero.
Rubrik provides scalable backup with automated setup and predictive search for granular restores. Rubrik's cloud-native backup works across multiple platforms, including AWS, Google Cloud and Microsoft Azure.
SolarWinds focuses on protecting both physical and virtual servers across all major OSes and hypervisors. Deduplication and compression functions help to accelerate backups. A local SpeedVault cache option is designed to speed recovery.
Veeam offers S3-compatible object storage, and a scalable and easy way to integrate with public cloud platforms.
Velero, which is backed by VMware, is an open source offering that combines standard backup and restore functions with the ability to migrate Kubernetes cluster resources and persistent volumes.