Taking a backup is the easy part.
Getting it back is the product.
Everyone has backups. Far fewer have ever tested a restore. GateCore finds every machine you own, shows you which ones nothing is protecting, and lets you open a virtual machine's disk and pull out a single file without restoring the machine at all.
You can't protect what nobody remembered to add.
Most backup gaps aren't technical failures. They're a machine somebody built in a hurry two years ago that never made it into a job. Connect GateCore, and the inventory builds itself — hosts, the virtual machines on them, and every physical computer.
| Name | State | Gen | vCPU | Memory | Size | Disks | Protected by |
|---|---|---|---|---|---|---|---|
| APP-01 | Running | Gen 2 | 12 | 31 GB | 90 GB | 1 | All Servers — Contoso |
| BACKUP-SRV | Running | Gen 2 | 16 | 20 GB | 99 GB | 1 | Not protected |
| FILE-SRV | Running | Gen 2 | 16 | 7.9 GB | 276 GB | 1 | All Servers — Contoso |
Click a row, and the actions come to you
What it is, what protects it, and the three things you'd most likely want to do next — without hunting through a menu.
Unprotected is amber, not grey
Grey means unimportant. A machine nothing is backing up is the opposite of unimportant — so it's the colour you can't ignore.
Decommissioned things stop counting
A machine that genuinely no longer exists can be forgotten, instead of sitting in the list looking like a failure forever.
Five steps, and then you stop thinking about it.
No backup server to build, no proxy to size, no afternoon lost to an installer. Pick the machines, say what to take, choose where it goes, set a time.
New job
NameAdd object
Pick a whole host, or single machines inside itTick the host and you're done.
Tick one machine and only that one goes.
Two ways of working, and neither one is a compromise. Take the whole host when you want everything on it covered without thinking about it again. Take individual machines when only three of the fourteen actually matter.
Whole host — including what doesn't exist yet
Somebody spins up a new VM next March. It's already in the job. Nobody has to remember, which is exactly how machines end up unprotected.
Or name the machines yourself
A host full of test VMs where only the database matters? Tick that one. You see each machine's size before you commit, so you know what you're taking on.
And it tells you what it's about to do
Pick a machine on a host that isn't in the job yet and it says so — "also adds this host to the job". No quiet side effects you discover at renewal time.
Mix both in one job
One host whole, plus two machines from somewhere else. It's one job, one schedule, one destination — not three jobs to keep in step.
No agent inside the virtual machine
Nothing to install on machines you may not even have a login for. The host does the work, so a VM you inherited is protected the same day you find it.
Files that are open right now
A database mid-write, a spreadsheet left open on Friday, a locked log. Captured as they stood when the job started.
Leave out what you don't need
Temp files, ISO images, folders that regenerate themselves. Stop paying to store noise every single night.
Keep this morning. And last March.
Daily copies for the mistake somebody made an hour ago. Monthly copies for the invoice dispute nobody saw coming. A yearly copy for the auditor. You set the shape per job — and old points are cleaned up on their own, so storage doesn't quietly grow forever.
Run it now if you can't wait
About to do something risky? Trigger a job by hand, on one machine or all of them — even while the schedule is paused.
Pause without losing history
Migrating a server this weekend? Pause the job, keep every restore point, switch it back on Monday.
Don't flatten the office line
Cap how fast a job uploads. The first full backup of a file server won't take the internet down while people are working.
Your data, wherever you want it kept.
GateCore Cloud, your own S3 bucket, or a share on your own network. Use one, or use all three at once — and every destination is tested, so an unreachable NAS surfaces before the day you need it.
Modern ransomware looks for your backups first.
It knows encrypting your servers is only leverage if you can't just restore. So it hunts down the backup share, the mapped drive, the credentials sitting on the backup server — and destroys those before it announces itself.
Encrypted before it leaves the machine
Every backup is encrypted on the device itself. What travels, and what sits in the vault, is already unreadable.
Append-only — written, never wiped
GateCore Cloud accepts new backups but won't let anything already written be removed. A machine that has been taken over can add to your history. It cannot erase it.
The password isn't on the machine
Each repository's encryption password is generated on the server, never displayed, never leaves it unencrypted. There is nothing on the endpoint for an attacker to steal.
New repository
Seven servers. Nowhere near seven servers' worth of storage.
Your machines are mostly the same machine — the same Windows, the same updates, the same application installed seven times. Identical data is stored once, no matter how many machines it came from, and what's left is compressed before it's written.
Stored once, referenced everywhere
Twenty machines running the same Windows build don't cost you twenty copies of Windows.
Yesterday's backup doesn't double it
Each new restore point stores what actually changed. Keeping thirty days doesn't mean thirty times the space.
Which is why the price works
Storage is the part that usually bites. Less of it means longer retention on the same bill.
Nobody notices a backup that works.
So the one that didn't is impossible to miss.
Jobs run at three in the morning while everybody's asleep. The next morning there's one line that matters — and if it isn't green, you know before a customer tells you.
Pick a machine. Pick a moment.
Then decide how much you want back.
Restore points are grouped by day, so you're choosing "last Tuesday morning", not reading a wall of timestamps.
Browsing a 126 GB disk costs you megabytes.
This is the part most backup tools make you pay for twice — once in time, once in disk space. GateCore opens the machine's disk read-only, straight from the vault. Nothing is downloaded up front. You walk the folders, tick what you want, and only those bytes ever leave.
It looks exactly like the machine did
Program Files, Users, your app's data folder. If you'd know where it lives on the server, you know where it lives here.
Tick a folder, get everything inside it
One file, one directory, or a handful scattered across the disk. The size of the problem decides, not the tooling.
Land it wherever you like
Back where it came from, or somewhere safe to check it first. Files already there are never quietly overwritten — you're told, by name.
Restore files
Backup shouldn't mean another product to learn.
The machine you're restoring is already in this console. You know its security score, who signed into it, and you can connect to it. Now you can get its data back too.
Protection
Activity
Access
One price for the software.
Storage only if you want ours.
Protection is a flat monthly figure, whatever you're backing up. Storage is separate, because plenty of businesses already have somewhere to put it.
On the roadmap
Virtual machine protection covers Hyper-V today. VMware and Proxmox are being worked on — we'd rather tell you that plainly than let you find out after you've bought it.
The restore is the product.
Come and test ours.
GateCore Backup is a Business add-on — it joins the console you already use.