Your final line of defense against the day everything else goes wrong.
Don't trust the survival of your company to a single credential boundary. Twin Vault is an independent, logically air-gapped recovery layer built beneath Veeam. We give engineers the technical tools they need, and executives the verified proof they demand: when a disaster or ransomware attack hits, the business will come back online.
Cheap storage is a budget line item. Verified recovery is a business survival strategy.
Every company starts the same way: a spare server and a basic backup target. But when a breach happens, you aren't accountable for the cost of the hard drives — you're accountable for the cost of downtime. Twin Vault is engineered for organizations that refuse to guess how long it will take to resume operations.
Because the recovery system needs protection too. We place storage-level recovery points beneath Veeam administration, protected by separate repository controls. Compromising the backup console should not give an attacker the only keys that matter.
Because a building is a failure domain. The second repository lives on separate infrastructure with separate credentials. A fire, flood, theft, or destructive attack should not become the end of the business.
Because a green check mark is not a restore. We test recovery on a schedule, document the result, and fix what fails. Evidence first. Confidence second. Recovery when it counts.
Predictable pricing based on your exact risk profile.
Tell us your required time-to-recovery (RTO) and data footprint. We build the exact architecture to match your business requirements — whether you need budget-friendly spinning disk or instant-recovery all-flash arrays. Purchase via outright CapEx or finance it as an OpEx lease. No hostage situations, no hidden licensing ceilings.
| Model | Chassis | Drives / layout | Usable capacity* | Starting price (CapEx)† | OpEx & financing | Best for |
|---|---|---|---|---|---|---|
| TNV2 | Dell PowerEdge 1U | 2 × 3.5″ ZFS mirror | 6–16 TB | $1,754 | Lease-to-own available | Single-server shops, entry immutability |
| TNV4 | Dell PowerEdge 1U | 2 × 2-drive ZFS mirrors (or 4-drive RAIDZ1) | 13–38 TB | $3,152 | Lease-to-own available | Growing SMBs & branch offices |
| TNV6 Popular | Dell PowerEdge 2U | 6 × 3.5″ RAIDZ2 | 38–102 TB | $5,025 | Lease-to-own available | Multi-server enterprise environments |
| TNV12 | Dell PowerEdge 2U | 12 × 3.5″ — 2 × 6-drive RAIDZ2 vdevs | 77–205 TB | $11,901 | Lease-to-own available | Large virtual estates, long retention |
| TNV18 | Dell PowerEdge 2U | 18 × 3.5″ (front + mid + rear) — 3 × 6-drive RAIDZ2 vdevs | 192–307 TB | $28,566 | Lease-to-own available | Enterprise capacity, multi-shelf expandable |
| TNV24F | Dell PowerEdge 2U all-flash | 24 × 2.5″ SSD — 4 × 6-drive RAIDZ2 vdevs | 49–197 TB | $15,900 | Lease-to-own available | Near-zero RTO — boot VMs live during a disaster; closes tight backup windows |
*Planning capacity is approximate after parity and a 20% operating reserve; final sizing also depends on change rate, compression, backup schedule, retention, snapshots, growth, and recovery workspace. TNV6→TNV18 grow in place, one 6-drive vdev at a time (plus a bay kit at 18) — a ten-minute zpool add, no rebuild, no forklift. Past a full chassis, the same pool extends with 12-bay PowerVault expansion shelves, two more vdevs per shelf — designed and validated, not guessed from a web calculator. Why 6-drive RAIDZ2 modules? The smallest storage module that survives two simultaneous drive failures while keeping two-thirds of its drives as data, and a failure rebuilds six drives — not eighteen. Every model is built from the same proven module.
On the all-flash tier: when the backup window is the constraint — hourly restore points, or nightly jobs that keep overrunning as merges, synthetic fulls, and a dozen jobs land on spinning disk at once — flash is what closes it. We run this architecture ourselves: all-flash repositories land the backups in under an hour, then a Veeam backup copy job moves them to a spinning-disk TNV for long retention on site. Fast landing zone, deep history — each on the media built for it. And restores read at flash speed: boot VMs directly off the repository while you rebuild.
- Preserve cash flow: lease-to-own structures are available through our B2B financing partners.
- No hidden egress or licensing fees: you are financing a physical, owned asset — not paying a variable tax to access your own data.
- Streamlined approval: financing approval is handled directly during the design review, alongside your quote.
Final monthly rates are calculated during the written design phase based on your exact hardware tier, support level, and current terms.
†Starting appliance prices are planning figures, shown as outright purchase. Drive-market changes, freight, tax, exact warranty, support tier, and final design can change the quote. Managed monitoring is optional at $199–$399/mo. Twin Offsite Vault planning estimates begin around $12/TB/mo for one region or $21/TB/mo for two regions; billing and retention are confirmed in the written design.
Two independent systems. One air-gapped safety net.
We enforce true separation of duties. Your backup software (Veeam) manages jobs, Fast Clone, and daily application restores on hardened XFS. Beneath it, an entirely separate storage layer (OpenZFS) handles checksum integrity, self-healing, and block-level offsite replication. Because they use entirely separate administrative credentials, a threat actor who compromises one does not hold the keys to the other.
ZFS recovery points are an additional operational control—not a substitute for Veeam immutability or certified WORM where regulation requires it.
Why these two file systems? XFS is enterprise Linux's workhorse — the default in Red Hat Enterprise Linux, and the file system Veeam speaks natively. ZFS was built at Sun Microsystems and has protected enterprise storage for twenty years. Veeam gets the repository it expects; you get an independent recovery layer beneath it.
Separate recovery points
Veeam writes to hardened XFS. ZFS snapshots are created and retained through a separate administrative path. When those controls and credentials remain separated as designed, compromise of the Veeam console alone does not provide the storage-layer deletion path. Every block is checksummed, and redundant layouts can repair detected corruption from a valid copy.
Replicate the repository
The repository can replicate block-by-block to separate infrastructure with separate credentials. Recovery points and backup history move together. You are not copying random backup files into a bucket and hoping someone can reconstruct the environment later.
Prove the restore
A completed job proves that data was written. A tested restore proves that it can come back. We test on a schedule, document the result, and remediate failures. That evidence exists before the incident, not after it.
Fast recovery for daily hiccups. Survival for worst-case scenarios.
Most downtime isn't a smoking crater — it's a corrupted database, a failed patch, or ransomware caught early. Twin Vault recovers these everyday emergencies locally at LAN speed on hardware you own, minimizing operational disruption. Your offsite, air-gapped copy is reserved strictly for the day the building or the network is completely gone.
Local-first — the everyday case
Deleted file, a bad VM, ransomware caught early? You restore from the on-site immutable repo at LAN speed. You never touch the cloud. It's fast because it's local — on hardware you own.
Full-site disaster — box & site gone
This is when the offsite copy matters. Depending on the contracted recovery design, dataset size, and destination, encrypted physical-media transfer or a temporary Twin colo recovery environment can be arranged. The written plan identifies the available path, dependencies, and target—not a fantasy RTO.
Instant recovery — TNV24F
On the all-flash tier, boot and run your VMs directly off the repository while the environment rebuilds. Instant virtualization on hardware you own — production keeps moving until you're ready to fail back.
No fantasy RTOs. Dataset size and available bandwidth still matter. We design around that reality: local recovery for common failures, separate infrastructure for site loss, physical seeding when the wire is too slow, and flash where workloads must run immediately. Physics is real. The plan should be real too.
We didn't set out to build another storage product.
We're an MSP. We've spent years designing, buying, managing, and recovering real customer infrastructure — including more than $100,000 in commercial storage software and licensing. Eventually we started asking a simple question:
What are we actually paying for — and what do our customers actually need?
We didn't need another storage interface. We needed recovery infrastructure we could trust and operate: hardened Linux, XFS, OpenZFS, encryption, independent recovery points, offsite replication, enterprise hardware, security monitoring, restore testing, and alerts that reach the people responsible for acting on them.
We wanted it built on technologies we understood, on hardware our customers could own, without putting another proprietary storage platform underneath their backup software. So we built it for ourselves. Then we realized other IT teams were trying to solve the same problem.
Twin Vault is the recovery infrastructure we wanted for our own customers.
— Chris Brown
Founder, Twin Networks
Own the first recovery layer. Add the second before you need it.
The local repository gives you fast recovery and storage-level protection. Monitoring makes failure visible. Offsite replication removes the building, the local network, and the local credentials from your final line of defense. Two is one. One is none.
Own the repository
Buy outright or finance lease-to-own. The immutable repository is yours — hardware, ZFS-backed storage, and Veeam-hardened configuration. No recurring cost required to be immutable locally. Managed monitoring is optional.
Managed monitoring
We watch the box and the backups — dial-home hardware alerts, OS patching, and managed Veeam job monitoring — even on a repository you own outright. A failed job or failing drive becomes our alert, not your Monday surprise.
Offsite replication
A separately controlled zfs send replica of the repository, with retained storage-level recovery points—one region, or two-region geography where the risk or compliance program calls for it. This design supports insurer and examiner evidence; it does not replace a control-specific compliance review.
Our guarantee — and what onboarding includes
Operationally immutable or it's free · acceptance verified by a documented restore test. The final quote defines the acceptance test, exclusions, and onboarding scope—including the Backup Exposure Assessment, migration plan, recovery playbook, and acceptance restore test. We do not call it done until the agreed restore test succeeds.
Repository-Level Recovery — add-on
Retained ZFS snapshots create independent recovery points for the repository itself. Mount a known-good snapshot and recover selected data without disturbing the active repository—or, when appropriate, restore the repository to a retained point in time.
Prefer one predictable number? The 3-year Managed Repository bundle.
Hardware, managed monitoring, and offsite replication in a single monthly payment over 36 months — no capital outlay, no separate invoices. Built for MSP-style, budget-friendly delivery to your firm.
The repository doesn't operate alone. A recovery system needs someone watching it.
Managed deployments include centralized infrastructure monitoring, hardware health, Veeam job monitoring, OS lifecycle management, security monitoring, replication monitoring, and scheduled restore validation. When something fails, the objective isn't another dashboard turning red — the objective is for the problem to become someone's responsibility. Hardware, storage, backup, security, and recovery events feed into Twin Networks' service workflow so they're investigated and remediated before the repository is needed during an incident. A failing drive becomes our problem. A failed backup becomes our problem. A replication failure becomes our problem.
Dial-home hardware monitoring — included
Every TNV reports its own health home automatically—drive SMART status, PSU and thermal state, and iDRAC9 hardware alerts. A failing disk becomes a planned service event under the warranty and response terms stated in your quote, not an outage you discover at 2am.
Managed monitoring & optional active defense — add-on
We watch the jobs, not just the box — job success, retention health, and repository capacity, backed by our team. On Twin-managed Linux configurations where the design supports third-party security tooling, the repository can add commercial EDR with 24×7 SOC monitoring, so a tampering attempt on your last line of defense is detected and responded to, not discovered later. This is a deliberate part of the Twin-managed Linux operating model — distinct from Veeam's own managed hardened-repository appliance, which follows a different security and software-support model. You never get the call that the backup silently stopped running three weeks ago.
Managed monitoring + EDR & 24/7 SOCVerified restore reporting
Scheduled restore tests with written results, so recoverability is a documented state — the evidence underwriters and examiners want to see — not an annual fire drill.
Security assessment — upsell
Want the full picture around the repository? Add a Twin Networks security assessment — backup posture, immutability gaps, and insurer-questionnaire readiness, reviewed by a practitioner who's answered those questionnaires for regulated firms.
Optional advisoryOwn your recovery. Don't rent it.
Twin Vault isn't another storage platform. It's a managed recovery architecture built from open, proven infrastructure. Neither model is wrong — but they are very different bets. Here is the difference in practice:
Proprietary Appliances
The rental modelWhen you outgrow the box, the path forward is a bigger box — a new purchase, a data migration, and a weekend you won't get back.
On many platforms, access to your own backup history rides on an active license. Worth asking your vendor exactly what happens to the data the day you stop paying.
The operating system and storage internals are off-limits. You can't inspect how corruption is detected or handled — you take the vendor's word for it.
Twin Vault
The ownership modelExpand capacity by simply adding a shelf (JBOD). Run a 10-minute command — no data migration required.
Zero capacity licensing. Standard Linux file systems (XFS/ZFS) mean your data is always yours.
Built on OpenZFS. Engineers can verify checksums, self-healing, and repository integrity.
Where a purpose-built appliance may win—and our answer. If a specific rule, record class, insurer, or contract requires certified Compliance-mode WORM, we will say so. The design may pair the fast, open, owned TNV repository on-prem with a qualifying S3 Object-Lock tier for the records that require it. Where a program requires FIPS-validated cryptography, that is a specific design we build and document explicitly — it is stated in the written design when it applies, never assumed from this page.
Built for the team that owns the recovery.
Twin Vault is designed for IT teams, MSPs, and organizations already running Veeam that want local recovery performance, hardened repository immutability, an independent storage-level recovery layer, offsite recovery under separate infrastructure and credentials, infrastructure they can own and move, and monitoring with restore validation — without becoming Linux or ZFS experts.
If all you need is the cheapest place to put backup files, there are less expensive ways to do that.
If you're accountable for whether the business comes back — let's talk.
Every repository starts with a design, not a checkout.
Configure it online. Have an engineer validate it. Use the configurator to estimate capacity, hardware, recovery options, and price — then, before anything is ordered, Twin Networks reviews the configuration against your production capacity, retention, change rate, recovery requirements, and growth assumptions, and sends the final design and quote for approval. Own it outright, finance it through our leasing partner, or have Twin manage it end to end. Infrastructure this important deserves a human sanity check.
Size My Repository
Use the configurator — enter your production data, get the build and the number.
Size & ConfigureGet a Repository Design
A practitioner sizes it with you and sends a straight quote — no obligation.
Request a DesignEvery repository is individually sized, built, hardened, documented, and acceptance-tested by Twin Networks. Pricing is starting-at, hardware only — final list may shift with the drive market. Ask us for a current quote.