Introduction to Next Generation Cloud Backup with Rubrik

    It is critical for enterprise IT to have a working and reliable backup solution for the data centre, recoverability time for any business can have a huge impact on costs.  For years now backup infrastructure has been difficult to implement efficiently and can require a lot of time to initially configure and to manage.

    With virtualisation the data centre moved from a physical server for each application to a more condensed and efficient footprint running VMs and with it, agentless backups that could target the hypervisor direct became popular and made backups easier to setup and manage with slick new interfaces.  The supporting infrastructure for these produces still remained though and proved time consuming to design and manage, especially at scale.

    Components to the infrastructure would include management servers, backup and replication proxies, backup repositories, external file systems or shares, databases and some sort of off-site archive function.  At scale this can become a big overhead to manage.

    A new generation backup solution – Rubrik aims to take away the overhead and make data management easier for the modern data centre.  Rubrik consolidates the same infrastructure functions into one converged appliance, known as a Brik, which can be a physical appliance or a virtual one.  A Brik appliance comes in 2U appliance made up of up to four x86 nodes, the solution is easily scaled by simply adding more Brik appliances.

    Rubrik continues its next gen capabilities by focusing on a API driven architecture, publishing the RESTful API to the admin to consume as well as partner integration with the likes of VMware vRealize Automation.  Rubrik software consumes the same published API as the admin would, ensuring reliability.

    Automation is the key to a successful software defined data centre and the Rubrik platform fits perfect into this solution.

    Features

    • Policy Driven - Protection policies are used to define protection for workloads by a given Recovery Point Objective (RPO), retention period, replication settings and if any archive is required, rather than managing backup jobs for applications.
    • Archive - Data can be archived from the on-premises device to cloud based storage such as AWS S3, this can be used for long-term retention outside of the Rubrik appliance.  This will require a cloud subscription however.  On-premises object based devices can also be used if sending the data to a public cloud is not a variable object.  Either way it’s important to consider archiving, general good practice would be to have multiple copies of the data and on multiple platforms.
    • Replication - Multiple Rubrik appliances can replicate data between sites if required with required RPOs and retention managed through policies. Rubrik can be run as a cluster in a public cloud and can run on multiple public clouds, such as AWS and Azure, enabling cross cloud replication should it be required.
    • Instant Recovery - Instance recovery is available, meaning a workload can be instantly mounted on the Rubrik providing near-zero Recovery Time Objects (RTO).  Instance recovery can also be leveraged in the cloud.  A new VM is created using the backed up data and powered on, replacing the original VM, running on the compute within the original cluster and storage running on Rubrik, the workload is running in production in a very short time and can be migrated back onto production storage live.
    • App Mobility - If the data is replicated to public clouds such as AWS, the data can be used and converted to run the cloud.  Amazon Machine Images (AMI) can be created and then run within the cloud, allowing for easy application mobility.
    • Global search to restore files and folders - Predictive global search unifies data locked within disparate application silos into one globally indexed namespace. File level restores can be delivered from any local snapshot, archive snapshot or replica that has been indexed.  To successfully index the VMware tools must be running on the protected VM, once indexed files can be searched and recovered from the Rubrik User Interface.

    Security

    Backing up data is critical and that data needs to also be very secure, recent ransomware attacks such as ‘WannaCry’ have highlighted this more than ever.  Imagine a scenario where production systems are affected by a ransomware attack and the backup server was also attacked, the data required to restore the production system could potentially be locked away.

    Rubrik backups are stored in an immutable fashion, never available in a Read/Write state to the client, the underlying backups remain in a Read-Only state preventing a ransomware attack affecting the backup data.  If a protected VM is deleted in production the backed up data is not deleted, instead it is aged out given the policy configured.

    Depending on the model of Rubrik, encryption at rest can be enabled in either software or hardware.  Software encryption at rest is provided via AES-256 encryption whilst hardware encryption at rest is provided compliant self-encrypting drives.

    Architecture

    • Fault Tolerant - Using multiple nodes in a cluster ensures the system and data availability should a node fail or a disk drive fail.  Replication is taken care of to ensure multiple copies of the data are made throughout the cluster.
    • Flash Optimised - The Rubrik system is built with a hybrid approach, using a flash/disk architecture.  Ingress data to the Rubrik is through the flash tier, processed through a number of parallel data streams to ensure maximum performance.
    • Global Dedupe - Data reduction is applied at a global level, ensuring efficiency through the system.  Global deduplication and compression is applied before the data is stored on the file system.
    • Easy Deployment - Due to the nature of the Rubrik appliance the deployment is very fast and simple, once racked a IP address is assigned to each node in the cluster.  Once deployed, simply add the managed environment such as an on-premises vSphere cluster and the environment is auto discovered utilising the VMware vStorage APIs for data protection.  Additional nodes can be automatically added to the cluster with an IP address and credentials.

    Conclusion

    Next generation backup solution from Rubrik complements the movement to a software defined data centre very well whilst also aiding the business’s public cloud strategy.  It bridges the gap between on-premises infrastructure and public cloud by decoupling data, protecting workloads no matter where they run.

    Rubrik focusses on simplicity, easy to scale, security and automation.  Recoverability is critical for every business, Rubrik make this very simple to initiate and manage whilst ensuring the solution can be easily scaled out and keeps the data secure.

    To find out more about the topics covered within this post or to discuss them with an Insight solutions specialist, please get in touch.
     

    Why not also read 'Data centre Automation and Operations'?