ESG Lab recently completed testing of the Rubrik Cloud Data Management Platform, which delivers backup and recovery, offsite replication, and data archival in a simple, scale-out appliance built for hybrid cloud. Testing focused on the simplicity of the Rubrik solution for managing and protecting virtualized and physical environments. ESG Lab examined how Rubrik users can get started quickly and easily, set up policies to automate backups, perform restores in a variety of ways, drive down recovery time objectives (RTOs) with live data access, and locate backed-up systems and data instantly with Google-like search.
Demands on organizations to manage data and storage requirements more efficiently continue to increase as environments become increasingly complex. ESG’s 2016 IT Spending Intentions Survey revealed that Improved Backup and Recovery is a top-five priority, and Business Continuity and Disaster Recovery is a top-ten priority, as they have been for years.1
As these areas of IT focus become more business-critical and data becomes more distributed, the need for integrated data protection, business continuity, and disaster recovery solutions increases.
The Solution: Rubrik Cloud Data Management
Rubrik Cloud Data Management is a software solution delivered in a turnkey appliance that horizontally scales across on-premises infrastructure and public clouds. Rubrik’s approach is a significant departure from the more typical backup and recovery architecture, in which backup software and protection storage are separately acquired, usually from different vendors. Its converged approach endeavors to unify backup, instant recovery, replication, search, analytics, archival, compliance, and copy data management into one appliance. The core design principle is to bring simple interfaces and true ease of use, reminiscent of Apple’s consumer products, to the enterprise data center.
Rubrik offers a flash-optimized web-scale infrastructure designed for backup and recovery, allowing users to horizontally scale a single solution to manage ever-increasing data volumes. “Web-scale” refers to the architecture pioneered by companies such as Google, Facebook, and Amazon to enable scaling performance and capacity by adding resources. The architecture was conceived to maintain resiliency through infrastructure failures and to be operationally efficient as the system grows in size.
Rubrik is engineered to automatically discover virtual machines without agents. It requires the one-time installation of software connectors to enable communications with physical MS-SQL and Linux servers. Rubrik can be configured to protect each type of machine at appropriate service levels with a single click, and can automatically archive snapshots offsite to the cloud to enable business continuity. Granular restores of individual files or application objects are simple and quick whether restoring locally or from the cloud, and they can be completed without administrative support. It is straightforward to live-mount snapshots onto virtual machines for restores, to run secondary workloads in test/dev, or to obtain analytics. Rubrik employs SLA Domains to manage protection policies across all of the machines that it manages. Once an SLA Domain has been created, an administrator can add one or more servers to it, and its protection policies are applied to those servers.
ESG Lab looked at how organizations can use Rubrik to rapidly deploy a data protection solution in a simulated enterprise environment, with a goal of easily deploying Rubrik and its agents, discovering and protecting systems, restoring backed up data, and seamlessly moving protected data offsite to the cloud.
ESG Lab used a four-node Rubrik cluster to protect a simulated data center environment consisting of physical machines running MS-SQL on Windows Server and file services on Linux as well as virtual machines running multiple applications in a VMware vCenter environment.
First, ESG Lab logged into the Rubrik web management console to configure the cluster, which required naming the cluster, entering the IP addresses for the nodes, configuring DNS and NTP, and clicking Setup System, as shown in Figure 3.
The Rubrik cluster was successfully set up in just over five minutes. Next, ESG Lab entered our VMware vCenter credentials to enable connectivity to virtual servers. Rubrik automatically discovered all virtual machines running VMware Tools in the cluster without the need for additional agents.
To add connectivity to Windows Servers running MS-SQL, an administrator must perform a one-time download of a lightweight software connector to the Windows Server. This can be accomplished either by hand, or with an orchestration tool such as Chef or Puppet to push the connector to a group of Windows Servers. Once the connector is in place, Rubrik will find the server, establish a connection to it, and add it to its list of Windows Servers, as shown in Figure 4. The connector will discover all instances of MS-SQL running on the server as well as any databases running within those instances. Rubrik supports physical and virtualized instances of MS-SQL Server versions 2008, 2008 R2, 2012, and 2014, running on Windows Server 2008 R2, 2012, and 2012 R2.
Next, ESG Lab created an SLA Domain. “SLA Domains,” in the Rubrik vernacular, define the frequency and retention period for snapshots as well as policies for onsite and cloud data retention and archiving. Rubrik offers preconfigured Gold, Silver, and Bronze SLA Domains, which differ in the frequency of snapshots, retention periods, and timing of archiving to the cloud. Users can also define their own SLA Domains to meet the requirements of their environment, as shown in Figure 5.
ESG Lab created a custom policy with daily snapshots retained for 30 days, monthly snapshots retained for one year, and annual snapshots retained for two years, plus the onsite data retention time was set to six months with the slider on the same screen. At this point, the cluster was ready to begin protecting systems with these customized policies. A single SLA Domain can be applied to both physical and virtual servers without modification.
The process of adding Rubrik to general-purpose Linux servers is very similar to the process for Windows Servers. Rubrik offers a software connector that can be pulled to each server manually, or it can be pushed using an orchestration tool. Once the connector has been installed, and is added to the Rubrik cluster, Rubrik traverses the entire directory tree within each server. Administrators can set up templates to manage the backup of directories that are the same across many servers. For example, an administrator may choose to back up local databases stored in /db/local directories every four hours, while backing up database executables and configuration files only once per day. The template can also exclude specific subdirectories or file types (e.g., /db/local/logs or *.log). Once a template has been created, Linux systems within the Rubrik cluster can be joined to it to ensure that their file systems are backed up according to the template rules, as shown in Figure 6. The Linux connector is compatible with versions 5, 6, and 7 of both Red Hat Enterprise Linux and CentOS.
ESG Lab then added a server to an SLA Domain. Finding connected machines was easy using the simple search interface reminiscent of Google Search. As the first characters of the server name were entered, Rubrik’s predictive search instantly listed matching system names, allowing for easy and rapid selection. When the name of the desired server appeared, a single click selected it. It was just as easy to select physical Windows and Linux machines as virtual servers and assign them to any SLA Domain as shown in Figure 7. Changing a server’s SLA Domain assignment was similarly straightforward.
Another aspect of backing up and protecting files is archiving them for longer-term protection and recovery. SLA Domains can be configured to enable automatic, scheduled archiving of backup data with a simple slider. Rubrik offers several options for archiving destinations:
- Amazon S3. Within Amazon S3, archives can be stored in three different Storage Classes, namely Standard, Standard – Infrequent Access, and Reduced Redundancy.
- Microsoft Azure. To store in Azure, the configuration must be initially performed on Azure to identify if a particular storage region is hot or cold, and then Rubrik can be configured to point to the storage region that’s been configured.
- Object Store. Archives can also be stored in several third-party object stores. Rubrik currently supports Basho, Cleversafe, Cloudian, IIJ, HDS, and Scality.
- NFS. Archives can be stored on an NFS server locally or remotely.
Why This Matters
Backup and BC/DR continue to be high IT priorities2 year after year for organizations of all sizes, due to the changing complexity and heterogeneity of production environments, as well as missed SLAs and dissatisfaction with traditional backup approaches. Data protection must evolve to meet these requirements simply and efficiently by delivering dynamic data protection management solutions that easily scale as the environments that these solutions protect grow and change.
ESG Lab was able to quickly and easily deploy a Rubrik data protection cluster and was protecting machines and application data, both locally and in the cloud, within five minutes of sitting down at the keyboard. What’s more, defining protection attributes was made effortless because Rubrik has combined traditional protection concepts into its simple SLA-based protection approach. The solution can be quickly and seamlessly expanded locally and in the cloud to address changing business requirements and data growth.
Restoring Rubrik Backups
Having successfully implemented SLA Domains and backups for the protected servers, ESG Lab moved on to restoring data by navigating to the Rubrik dashboard shown in Figure 8, which presents restore options as well as system, performance, and utilization information in an easy-to-read interface. Searching for machines with the Google-like interface was as simple as described above.
There are several options for restoring files in Rubrik. A user can browse complete backups either through a list of servers, or through a calendar interface to find the date and time required. With a Linux backup, a user can restore a whole directory or an individual file from any backup set that is still available on the Rubrik cluster. An example of this is shown in Figure 9. In that example, since backups of /etc are scheduled every four hours, the user is presented with six backups, taken at four-hour intervals over a 24-hour period.
ESG Lab chose to restore a single file, /etc/aliases, from a backup. As shown in Figure 10, it required a search for snapshots of all files whose name contains “alias.” The figure shows that there are 30 stored versions of /etc/aliases available. One click on the “30” returned a list of all 30 versions, from which it was simple to select the desired version. After the version was chosen, the next screen presented several options regarding where and how the backup will be restored:
- Restore in Place. This is a traditional restore, where the restored files replace what is currently active.
- Restore to a Different Folder. The user can choose to restore the contents of a directory or specific individual files to a different directory on the same server.
- Export. An export allows the restore to be done on a different server. This can be useful for creating a clone of the originally backed up server, setting up a test/dev environment, reproducing an existing server without disturbing it, or migrating from a physical server to a virtual one, all without impacting production operations.
- Download. The contents of the backup can be downloaded into a single compressed file.
Rubrik can also restore a database from its snapshots. ESG Lab logged in to the Rubrik dashboard, selected PROTECTION > SQL Databases from the left-side menu, and chose the correct server. That brought up the screen shown in Figure 11. On the left side of the screen was an overview of the protection options in place for that server, and on the right side was the Recovery Points screen. Rubrik offered a variety of calendar options to help find the desired snapshot: the whole year, one month, or a single day, and within the single day option, there was a slider to help select the exact time of the snapshot. ESG Lab chose to restore the snapshot taken on August 1, 2016 at 6:30am. The last question was whether to restore it in place or on another system in the cluster. Another system was selected, and the restore began. No hunting for offline media or searching for the right backups was required. The interface is clean and easy to use.
A long-standing issue with restores is that users are typically required to ask their administrators to perform them, and as a result, the users are forced to wait until the administrator has time to complete it. That model creates extra work for the backup administrator and frustrating delays for the end-user. Rubrik allows users to log in to its interface and perform the restores themselves. As part of Rubrik’s design around ease of use, it offers Role Based Access Control (RBAC) functionality. RBAC allows individual users to log in to the Rubrik UI, but limits their access to the systems and backups that the Rubrik administrator has granted them; they can only see and restore the systems and backups that they are permitted to see, and they can perform those restores themselves.
ESG Lab created an end-user account with very limited access and logged in as that user. Instead of showing several different hosts, databases, VMs, etc., along with a complete Rubrik dashboard, the dashboard contained just a single MS-SQL server, as shown in Figure 12. This is a valuable security feature that will be welcomed by many shops and required in others. It can reduce the burden on backup administrators, who will no longer be called upon to perform these basic tasks.
Why This Matters
In addition to backup and recovery, managing data growth, and virtualization, nearly one in five survey respondents cited business continuity and disaster recovery when asked to name their organization’s most important IT priorities.3 As environments increase in complexity, the ability to leverage simple, effective tools increases in importance.
While Rubrik’s approach to data management and recovery is not traditional, it leverages its intimate knowledge and tight integration with Linux, MS-SQL, and VMware to provide radically simple, powerful offsite protection and recovery of modern data centers with a combination of physical and virtualized systems. Rubrik’s web-scale architecture delivers an elastic system that allows users to easily manage data growth by scaling out capacity and performance with the addition of nodes to the cluster. The system is managed as a single system from one interface, even as the system increases in size and capacity.
ESG Lab was able to use simple yet powerful search tools to find backed up systems, as well as individual objects, and recover them for instant restores, test and development, or any other purpose. In ESG Lab testing, Rubrik delivered a comprehensive data management solution that transformed backup data into a multifaceted business asset for business continuity and application development.
The Bigger Truth
As managed environments increase in complexity, the demands on IT organizations to efficiently manage their data and storage continue to grow. For the seventh year in a row, ESG’s 2016 IT Spending Intentions Survey show the high prioritization to improve both backup and BC/DR amid the backdrop of ever present challenges in server virtualization, improved backup and recovery, data growth management, and the desire to leverage cloud services.4 The increasingly distributed nature of the data along with its ever-growing business-criticality mean that the needs for integrated data protection, business continuity, and disaster recovery solutions continue to grow.
Organizations of all sizes are looking for a way to ensure that their applications and workloads are protected, simply and completely—whether on physical systems or virtualized. Rubrik Cloud Data Management is designed to automatically discover virtual machines without agents, protect all machines with a single click at the appropriate service level, and automatically move backup data offsite to the cloud for business continuance. At the same time, Rubrik enables simple, instant recovery of individual files and application objects—locally or from the cloud. While live mount is supported on virtual machines to run restores or secondary workloads, it is not yet available for physical systems. (Rubrik says that this feature is in development and scheduled for a future release.) Instant data delivery enables the creation of recovery plans—where users can easily and regularly test recovery without impacting production and have confidence that their data will be available for restore whenever they need it.
ESG Lab was able to set up a Rubrik cluster and begin protecting both virtual and physical systems with custom SLAs in less than 20 minutes. Recovering virtual machines was as easy as mounting a snapshot onto a new virtual machine. Entire MS-SQL databases were recovered from snapshots and individual Linux files and directories were selected and recovered after browsing a long list of backups taken at different times, regardless of whether the backups were stored locally or in the cloud, completely transparently.
The Rubrik Cloud Data Management Platform brings together backup software and deduplicated storage in a single fabric with the goal of simplifying data protection and management in environments with virtual machines, Linux, and MS-SQL servers. ESG Lab’s opinion is that Rubrik delivers outstanding simplicity and ease of use. Installation and configuration is simple and fast. Users can drive down RTO by live-mounting virtual machine backups, and instantly find data with Google-like search. Organizations that are looking to simplify data protection and business continuance would be well served to take a closer look at the Rubrik solution.
1. Source: ESG Research Report, 2016 IT Spending Intentions Survey, February 2016.↩
2. Source: ESG Research Report, 2016 IT Spending Intentions Survey, February 2016.↩
3. Source: ESG Research Report, 2016 IT Spending Intentions Survey, February 2016.↩
4. Source: ESG Research Report, 2016 IT Spending Intentions Survey, February 2016.↩