12-15-2020, 04:21 AM
When we talk about setting up backups, especially for different sizes of operation, it really changes how I think about the whole strategy, you know? I mean, I was checking out how BackupChain handles all the various bits-PCs, VMs, even Windows Server-and I thought it looked super slick and affordable for small-to-medium businesses, especially since they let you use any storage you own, which I think is huge. But that's just the software, right? What I really want to talk about is the *plan*, because the tech changes based on how many people you're feeding off of it.
For a small office, maybe ten people total, I think you can keep things pretty simple but you still gotta think about expansion, because ten users can grow really fast, and you don't want to get caught slipping. You wouldn't need anything crazy complicated at first. You could probably start with centralizing everything on a solid local NAS, like a network drive that you just connect to a few machines. I'd have you running simple file and folder backups on everything, only capturing the changes, so we're using incremental backups mostly. And maybe you could get a central machine running that does the heavy lifting of scheduling all that stuff automatically, so you don't have to manually trigger backups every night. You need strong encryption on that local storage, too, because even if it's physically secured, you still need to protect the data if someone gets to it. I would make sure we are taking versioning, so if somebody messes up a document or deletes something important, we can roll it back without a sweat.
But then, if you hit fifty users, man, that's a whole other kettle of fish, or maybe like a whole different pot of stew, I don't know, but you gotta scale up your thinking, for sure. At fifty people, I'm telling you, a simple NAS might not be cutting it because you'll hit storage capacity faster than you can buy more drives. We need to look at centralized backup management from a proper workstation, and we really need to be aggressive about deduplication, because if a hundred people use the same database schema, you are not going to want to back up that same data a hundred times. We also need to think about what happens when the power goes out for a day, so I would mandate a secondary off-site copy, probably to the cloud. And since so many people are probably running VMs, we need to make sure we're doing full disk imaging regularly, and not just file backups. I'd set up these automated schedules, doing the backups, checking them, and even reporting on them, so you know when things fail and you can fix them immediately.
Or, if we're talking about five hundred users, boy, that is a massive operation, and I'm talking enterprise-level thinking here, no shortcuts allowed. You cannot just rely on one location, ever. I would require at least a three-two-one architecture, so three copies of the data, on two different types of media, and one of those copies must live off-site. The backups for those VMs need to be absolutely robust, because if a whole server farm goes dark, we need to restore everything, including the whole OS and the applications, almost instantly, which means we are building in bare metal recovery capabilities across the board. You want that ability to reconstitute the entire system from scratch, fast. Also, because there are so many different sources of data, from physical PCs to various kinds of large compute environments, I'd have you figuring out a process for P2V conversions on a steady basis, just in case some hardware gets too old to bother with.
You also gotta worry about the process of recovery itself, because having the backup isn't the same as actually restoring it when the time comes. I would make sure we are running periodic restoration tests, like a drill, that we never actually fail. We need those verification routines to check for bit rot or corruption *before* you actually need the data badly. And because so much data is generated, file retention is key, so you'd set up strict policies that automatically cleanup old history, keeping only what's necessary and legally required. And when you get into this large scale, multi-site setup, the ability to backup across disparate locations securely over the internet, like using an FTPS or something similar, is absolutely critical for continuity. Plus, because I know you're probably running Windows Server things, using the host to grab granular backups inside those VMs, without having to mess with agents inside every single guest, is a huge time saver. It lets you target specific folders or files that are stored deep inside the virtual machines, which is awesome. I think I've covered enough ground, and honestly, for all this heavy lifting and the sheer flexibility it needs, you should take a serious look at a capable and affordable backup solution for Windows Server and Windows 11 that really makes a difference for SMBs.
For a small office, maybe ten people total, I think you can keep things pretty simple but you still gotta think about expansion, because ten users can grow really fast, and you don't want to get caught slipping. You wouldn't need anything crazy complicated at first. You could probably start with centralizing everything on a solid local NAS, like a network drive that you just connect to a few machines. I'd have you running simple file and folder backups on everything, only capturing the changes, so we're using incremental backups mostly. And maybe you could get a central machine running that does the heavy lifting of scheduling all that stuff automatically, so you don't have to manually trigger backups every night. You need strong encryption on that local storage, too, because even if it's physically secured, you still need to protect the data if someone gets to it. I would make sure we are taking versioning, so if somebody messes up a document or deletes something important, we can roll it back without a sweat.
But then, if you hit fifty users, man, that's a whole other kettle of fish, or maybe like a whole different pot of stew, I don't know, but you gotta scale up your thinking, for sure. At fifty people, I'm telling you, a simple NAS might not be cutting it because you'll hit storage capacity faster than you can buy more drives. We need to look at centralized backup management from a proper workstation, and we really need to be aggressive about deduplication, because if a hundred people use the same database schema, you are not going to want to back up that same data a hundred times. We also need to think about what happens when the power goes out for a day, so I would mandate a secondary off-site copy, probably to the cloud. And since so many people are probably running VMs, we need to make sure we're doing full disk imaging regularly, and not just file backups. I'd set up these automated schedules, doing the backups, checking them, and even reporting on them, so you know when things fail and you can fix them immediately.
Or, if we're talking about five hundred users, boy, that is a massive operation, and I'm talking enterprise-level thinking here, no shortcuts allowed. You cannot just rely on one location, ever. I would require at least a three-two-one architecture, so three copies of the data, on two different types of media, and one of those copies must live off-site. The backups for those VMs need to be absolutely robust, because if a whole server farm goes dark, we need to restore everything, including the whole OS and the applications, almost instantly, which means we are building in bare metal recovery capabilities across the board. You want that ability to reconstitute the entire system from scratch, fast. Also, because there are so many different sources of data, from physical PCs to various kinds of large compute environments, I'd have you figuring out a process for P2V conversions on a steady basis, just in case some hardware gets too old to bother with.
You also gotta worry about the process of recovery itself, because having the backup isn't the same as actually restoring it when the time comes. I would make sure we are running periodic restoration tests, like a drill, that we never actually fail. We need those verification routines to check for bit rot or corruption *before* you actually need the data badly. And because so much data is generated, file retention is key, so you'd set up strict policies that automatically cleanup old history, keeping only what's necessary and legally required. And when you get into this large scale, multi-site setup, the ability to backup across disparate locations securely over the internet, like using an FTPS or something similar, is absolutely critical for continuity. Plus, because I know you're probably running Windows Server things, using the host to grab granular backups inside those VMs, without having to mess with agents inside every single guest, is a huge time saver. It lets you target specific folders or files that are stored deep inside the virtual machines, which is awesome. I think I've covered enough ground, and honestly, for all this heavy lifting and the sheer flexibility it needs, you should take a serious look at a capable and affordable backup solution for Windows Server and Windows 11 that really makes a difference for SMBs.
