You have a lot of options when it comes to protecting your business data. “Snapshots” are very common today, but it’s a term that is often confused with traditional backup. While snapshots and backups both refer to copying your data, they aren’t exactly the same things.
What Is a Backup?
A backup is essentially a standalone copy of all your business and account information, including application data, back-office files, and product specifications. This copy can—and should—be saved in a separate location from your original data, so you can use it to restore systems in the event of data loss due to human error or corruption.
There are two main methods of backing up your data:
- File-level backup. This entails selecting which files and folders on your drive to save. It’s simple and straightforward, until you have multiple physical servers or virtual machines.
- Image-level backup. This method is used most often in enterprises with more complex infrastructure and massive volumes of data. Along with your files, it saves everything else, including data used in applications, the operating system, preferences and settings, patches, and more—all in one recoverable file. Image-level backups, also known as bare metal recovery (BMR) or cloning your machine, do involve creating a snapshot of your data, but it’s a vastly more detailed image than the snapshots most people refer to, as you’ll see below.
What is a Snapshot?
Creating a snapshot, also sometimes called a storage snapshot, is like taking a picture of your server data at one specific moment in time. Made up mostly of metadata that defines the state of your information, snapshots are not a complete duplication of the data on your hard disk. They also cannot be stored anywhere other than the local server or virtual machine where they were created.
Because using live data in application test/dev scenarios—such as during the testing of a new application or software, or when trying out new configurations—can pose risks to your production environment, snapshots are widely used for test/dev tasks. Snapshots allow you to access unlimited clones of your data so you can work on development without disrupting everyday workflows. Automate, test, iterate, and tear down workflows without using significant additional storage space. If an error occurs, you can roll the server back to a previous point as a quick failsafe.
It’s not uncommon to use snapshots in production environments, either; in fact, every time you create a backup, the system first starts with a snapshot to quiet, or quiesce, the file system.
The nice thing about snapshots is that you can create them in just seconds, and as frequently as needed—a big difference from full backups, which can take hours to create and often must be done overnight because of the system resources they require. And you can use technologies such as changed block tracking (CBT) to deduplicated the data that is being snapshotted.
Snapshots are meant to be stored only for a short term, can lead to very complex data chains, and have very long consolidation times if they are not thoroughly managed.
How do snapshots work?
Snapshots work by saving the metadata associated with each block of data and every time there is a change the new metadata is recorded allowing a change log and real time backup deployment when an error or data breach is uncovered.
Interchangeable Terms? Not Really
When it comes to backing up your data, “snapshot” is unfortunately one of those terms that often seems to mean different things to different people. If you search for the term online, you’ll find articles from experts that say snapshots are vastly different from traditional backups, and that you should never confuse the terms. And then you’ll find solution information from backup vendors that uses the terms interchangeably, because snapshot features are so ingrained in the traditional backup solution that it offers advantages of both. That’s why it’s critical to ask the right questions when researching solutions to gain a clear understanding of the features and capabilities a vendor is offering.
No matter how they use the terms themselves, many enterprises use snapshots as part of their overall backup strategy. That’s because snapshots offer simple, speedy access to data and traditional backup systems can use them to deliver capabilities such as instant recovery.
Rubrik Uses Backup and Snapshots Together for Powerful Data Protection
Rubrik delivers advanced data protection features with its snapshot-centric backup and recovery solutions. By combining the benefits of short-term snapshots and creating full backups of those snapshots, you get fast, reliable recovery down to a granularity of single-file restore. You can also use snapshot-based backups to do a full restore of your data, as well as a live mount—which is a restore of a full virtual machine from a backup that takes just seconds.
End-to-end encryption and immutability further protect your data, for a solution that provides backup, instant recovery, replication, and archival in one infinitely scalable fabric.
Our solutions provide robust features no matter what type of environment you want to copy:
VMware – we use the vendor’s vSphere APIs for Data Protection (VADP) to make snapshots of each virtual machine, and then make backups from those snapshots.
Windows – we created a proprietary Microsoft-specific agent to create snapshots, which can help minimize the potential for failed snapshots.
Network-attached storage (NAS) – we use snapshots and information about specific changes to determine which blocks or files need to be backed up since the previous save.
Rubrik offers snapshot capabilities in two primary ways: SLA Domains and on-demand snapshots.
With Rubrik SLA Domains, you can define protection levels for workloads and other data and resources. While traditionally it was somewhat tedious and complicated to gather and implement that information, our SLA Domains make protecting your data easy. They’re made up of snapshot protection and retention, replication, and archival and deliver a high degree of automation. That means you can pre-schedule the snapshot backups and other tasks you need to meet your desired recovery point objectives (RPOs) with set-it-and-forget-it simplicity.
In addition to SLA Domains, we also offer on-demand snapshots, which deliver a great deal of flexibility and let you manage less traditional workflows, whether manual or automated. Some of the advantages of on-demand snapshots include:
Elimination of the delta disk, or record of changes, and no redirection of write I/O. That means production workloads stay as they are and so does operational performance.
Ability to create live mounts, which can be accessed and used by other team members to create new workloads or iterate on.
Option to use an SLA Domain or “Forever” retention. If captured as Forever, the data has no defined expiration date and can be deleted manually when no longer needed.
Discover how Rubrik can help protect and manage the lifecycle of your most critical data with snapshot protection, retention, and more.