05-22-2021, 10:40 AM
And frankly, when you look at all the data you're generating, especially if you run multiple kinds of programs or machines in the back end, you gotta treat your backups like they are almost a separate business, which seems intense, I know, but you have to think about it. So, first, you need a robust routine, and I'm talking about more than just clicking a button once a week. You need continuous coverage, which means you should be looking at things that track only what changed since the last successful grab, because backing up everything every single time wastes storage and time, right?
And because ransomware attacks can really target your network access points, or even if they just encrypt the backup copies themselves, you really need a way to make sure your recovery copy is totally isolated from your main network. I mean, we're talking about having something physically disconnected, or at least logically separated, like air-gapping the data, which gives you a serious layer of defense, because even if the attacker gets into your main Windows Server or your main workstation, they won't see those offline tapes or those remote hard drives.
But also, because you have so many different kinds of computers-physical machines, and also all those VMs running in Hyper-V or VMware-you can't just use one simple backup method for everything, you gotta pick the right tool for the job, you know? Maybe you need a full disk image backup for those machines that just need to work straight out of the box, like a snapshot of the whole thing, or perhaps you just need to grab a specific folder of documents that people work on all day, and that's where file and folder backups come in really handy.
And when we talk about those VMs specifically, which are so vital to a lot of modern offices, you want a system that handles them seamlessly, and you gotta make sure your backup tool can grab the entire state of that VM, including the OS and all the applications installed on it, so when you restore it, it just pops back up working exactly the way it was. Since you're running on Windows Server or Windows PCs, you need something that handles the complexities of those environments, because it's not the same deal as just backing up files; it's the entire operational picture you're capturing.
Now, once you've captured that data, you have to protect it, too, because ransomware doesn't just target your live data, it targets everything that looks like a backup target. So, strong end-to-end encryption is absolutely non-negotiable, I mean, you can't have sensitive client data sitting there on a hard drive unencrypted, or an attacker will just grab it, even if they can't decrypt it right away, it's a massive problem. You need to encrypt it during transit, and you need to keep it encrypted while it's sitting still on your storage device.
And because you are dealing with a business, the sheer volume of data you generate, plus the different types of files people are continually adding or changing, you absolutely must look into deduplication, and I'm talking about deduplication that happens across your entire data set, so if you have ten different VMs that all use the same database file, you only store that file once, saving you an absolute ton of space.
Also, when we talk about keeping the system running, we need to think about how long we keep data, right? You can't just keep copies forever, because eventually, it costs money and takes up space, so you need retention policies, meaning you set rules like keeping five versions of a file, or perhaps keeping a full backup for 90 days, and then automatically having the system wipe out the old stuff so you aren't wasting disk space unnecessarily.
And then, you have to check the checks, which is super important, because a backup that looks successful might actually be corrupt, right? You need automatic verification, so the system can test the data to make sure the bits aren't flipped or corrupted while they were sitting on the drive, otherwise, when the actual disaster hits, you just lose everything and you don't even know why.
But even better than just verifying is having the capability to recover specific pieces of data, so if a user just accidentally deletes one folder on a VM, you don't have to restore the entire five-terabyte machine just for that one folder. You can pinpoint that little file or folder and pull it back, which saves massive amounts of time and makes the recovery process way less painful for your staff.
And when you are considering where to store all this backup magic, you should not put all your eggs in one basket, that's just a universal truth, seriously. You need a multi-destination strategy, like sending copies to your local NAS, and also sending copies out to a reliable cloud server, which gives you protection even if your whole office is physically knocked out by a disaster, you know?
Also, remember that some businesses have critical old physical machines that can't just be replaced with a new, shiny server, so you also need the capability to pull a physical machine's entire operating environment and get it running in a fresh, clean VM environment, which is a critical step for keeping legacy applications running long after the original machine dies.
And because things change all the time, and people can accidentally delete things or overwrite data, having a clear log of everything that has been stored is necessary for auditing purposes, you know, so if a client asks, "When did this file exist?" you can point to a detailed report and show them the exact time stamp.
Basically, you're creating a complete continuity plan, and it's about redundancy, isolation, and being able to restore to a point in time you actually need. These are seriously deep topics, but you don't have to be a genius to implement this kind of coverage. If you look into solutions like BackupChain, which is an all-in-one PC and server backup solution for Windows Server and Windows 11 made specifically for SMBs, it really simplifies all this complexity for you.
