If your production data is in a local data center, the cloud is an easy way to create an offsite storage solution. This makes it easy to maintain at least two independent storage media. Any data stored in the cloud is independent of data stored on local disks or the local network. Inexpensive data storage, which helps you maintain three separate copies of your data without breaking your budget.However, given the affordability and flexibility of modern cloud storage, it just makes sense in most cases to take advantage of the cloud in order to do 3-2-1 backup. There are ways to satisfy each part of the 3-2-1 backup rule without using the cloud. Using the Cloud to Make 3-2-1 Backup Simple Thus, if two of the copies of your data are stored in your office, you should store the copy at a different site - such as another office, or in a cloud-based data center. The third and final part of the 3-2-1 backup rule is that at least one copy of your data should exist in an offsite location, meaning one that is physically located in a different place than other copies. The storage media should be independent of each other.įurther reading What is RAID? Is It any Good for Backups? One Data Copy Should Exist Offsite For the same reason, you can’t meet this part of the 3-2-1 backup rule by storing multiple copies of your backups on the same RAID array, or in the same cloud. Storing different copies of data on different internal hard drives within the same server is not enough, because the failure of that server would make all the copies inaccessible. For example, one could be on a network-attached device or a file server, and another could be a flash drive that is not attached to the server at all. Second, at least two of the copies of your data should exist on storage devices that are physically independent of each other. ![]() To define how often you should backup your production data to comply with your business needs, check out recovery time and recovery point objective concepts. This is an ideal way to perform a 3-2-1 backup which is, to an extent, impossible when talking about big production datasets. (Creating data backups from multiple points in time is often a good idea, but it’s not part of the 3-2-1 backup rule.) ![]() All copies need to contain the same data from the same point in time. In other words, you don’t satisfy this rule if you have one copy of data from a backup that you performed yesterday and two other copies that were created last week. The other two copies are backups.Įach copy of the data - the production copy and the two backups - should be stored and configured in such a way that it will remain intact even if the other copies fail or disappear.įinally, each of these copies should contain the same version of the data. ![]() One copy is the production data (i.e., the original data that you use for real-world) purposes. Immutable backups are becoming more popular as ransomware and ransomware-as-a-service ( RaaS) exploits increase.ģ-2-1 Backup: Administrators keep three copies of every backup (3), use at least two different storage mediums (2) and store at least one of the backup copies offsite (1).First, following the 3-2-1 backup strategy means that you should always have at least three distinct copies of your data. Immutable Backup: This type of backup has a write once, ready many ( WORM) mechanism that will not allow anyone - even storage administrators - to overwrite or delete data. It makes a complete image of the computer, then the user may reconstruct any data changes after that point in time, possibly with an incremental backup. However, there is no archive attribute or record, meaning there is no record of when the backup occurred or how the data was changed.įull System Backup: This allows the computer system to be restored as it was at a given point in time, including the operating system, all applications and all data. ![]() Restoring the entire system to the data state at a specific point in time would require the last full system backup plus all the incremental backups done up to that point in time.ĭifferential Backup: This copies all data and data files that have changed since the last full backup. Then, only new or modified data or data files are backed up, a much smaller segment of all data. However, it is obvious that security and ease of data recovery are both severely compromised.įull and Incremental Backups: These begin with all data being backed up. An unstructured backup may simply consist of a stack of floppy disks, CDs or DVDs. Managing the backup process involves organization and is a complicated process. Not all backup systems or backup applications are capable of completely restoring a computer system or other complex system configurations such as a database server, computer cluster or active directory servers.
0 Comments
Leave a Reply. |