One operating system handles virtualization, storage, networking, and backup at once. Licensing is calculated "per physical server" - the price is the same no matter how many processors or cores are installed. We provide planning, deployment, migration, training, and annual maintenance.
VergeIO is a US software company headquartered in Ann Arbor, Michigan. It was formerly Yottabyte LLC, founded in 2010, later renamed and repositioned. It now makes a single product: VergeOS.
VergeOS is itself an operating system. It installs directly on bare metal (a physical server with no operating system installed). Once booted, this server is a virtualization host, storage node, network node, and backup node at the same time.
In plain terms: one operating system replaces the four or five licenses that would otherwise be bought, learned, and renewed separately. Virtualization software, software-defined storage, network virtualization, a centralized management console, plus a separate third-party backup product - VergeOS folds all of this into one codebase. One management interface, one licensing structure.
The vendor calls this approach UCI (Ultraconverged Infrastructure). The difference from typical hyperconverged infrastructure (HCI) is this: VergeOS builds storage into the operating system kernel, so each host no longer needs a memory-hungry controller virtual machine (CVM). Compute and storage can be scaled separately, and the same cluster can mix servers from different generations.
Former VMware CTO Kit Colbert and former ecosystem lead Zia Yusuf have invested and joined the board; the vendor has not disclosed the amount. Analyst firm DCIG placed VergeOS on its Top 5 VMware Alternatives list (SMB edition).
We source vendor products and support through authorized channels in Taiwan, and provide planning, deployment, migration, training, and annual maintenance services.
The following nine features are all included in the same VergeOS package, sharing one management interface and one licensing structure.
Pools the hard disks and SSDs from every server in the cluster into shared storage, replacing an external storage array (SAN). Storage logic is built into the kernel rather than routed through a controller virtual machine. It includes global deduplication and automatic tiering, and also supports connecting external storage.
Includes a virtual switch and router, BGP dynamic routing, firewall rules, DNS, traffic mirroring, and bandwidth throttling, and also provides isolation between tenants. It plays a role equivalent to VMware NSX, with no separate network virtualization license to buy. Note: VergeFabric only runs inside VergeOS.
Typical virtualization virtualizes "a single server"; VDC virtualizes "an entire data center." Each VDC has its own compute, storage, network, firewall, and administrator accounts - isolation happens at the infrastructure layer, not just the network layer. A whole VDC can be cloned, moved to another VergeOS instance, or nested. The vendor states explicitly that the number of VDCs is not charged extra.
Instant, independent, highly space-efficient snapshots that can be set as immutable. Before the retention period expires, they cannot be deleted even if an administrator account is compromised. Snapshots can be taken of a single virtual machine or an entire VDC.
Creates immutable, recoverable VDC checkpoints and detects ransomware behavior patterns; combined with VDC isolation, it reduces the chance of lateral spread after a single tenant is infected.
Replicates an entire VDC (including virtual machines, storage, and network settings) to another site. The key point is that network settings are restored along with everything else, so there is no need to rebuild the network at the disaster recovery site.
Built-in near-continuous backup, configurable at 15/30/60-minute intervals, transferring only deduplicated differential data. What sets it apart is how recovery works. If multiple disks or multiple hosts in a cluster fail at once, it feeds the missing data blocks to the still-running virtual machines in real time, so service is not interrupted. Data is backfilled once new disks are installed. The backup tier can use inexpensive HDDs.
Automates VMware virtual machine migration to VergeOS at scale. The vendor claims very short downtime; actual figures should be confirmed through POC testing. This is a key tool in the deployment process: without it, moving the production environment would rely on slow, manual scheduling.
Native GPU resource pooling; the vendor states no separate NVIDIA vGPU license is required. Includes a built-in OpenAI-compatible API router, supports deploying open-source large language models, and works in fully offline (air-gapped) environments. VergeIQ shipped with VergeOS 26 in October 2025.
What usually stalls a deployment decision is the risk of switching over the production environment - technology is secondary. ioProtect lets this decision be made in stages.
This is the recommended approach, and the path that keeps deployment risk to a minimum.
Enquire about an ioProtect disaster recovery site assessment View the hyperconverged infrastructure overviewVergeOS uses a subscription priced per physical server (per node). This is its most direct difference from "per-core" pricing.
The price is the same regardless of how many processors, cores, memory, or disks a machine has. The number of licenses equals the number of nodes - simple enough for procurement to verify on their own.
Under per-core pricing, moving to a new server with more cores pushes the license fee up. Per-node pricing does not. Taiwanese customers commonly buy high-core-count servers, so the gap between the two models is significant. This calculation can be worked out on the spot for procurement and finance during the evaluation stage.
The vendor states that the number of VDCs and global deduplication are included in the license. Snapshots (ioClone) and ioGuardian are also built-in VergeOS features. Whether centralized off-site backup management requires an additional license, and the actual license tier, term, and support level, all follow the vendor's current program. These are itemised in the quote.
You can supply your own hardware, use a VergeOS-certified appliance, or deploy on a rented bare-metal environment. Prices are not listed on this page. Actual cost depends on node count, term, and support level; please contact us for an assessment.
The following three scenarios are where VergeOS's architecture pays off most. The fourth item lists conditions where a different architecture is recommended instead.
Government agencies, defense supply chains, critical infrastructure, financial institutions, and Taiwan subsidiaries of foreign companies that must pass US supply chain audits. VergeIO's vendor is a US company; supply and support in Taiwan are obtained through the Asia-Pacific regional channel. If the review scope covers every tier of the supply chain, country-of-origin declarations and compliance documentation are prepared and confirmed during the evaluation stage.
Groups with multiple subsidiaries, business units that must be kept separate for compliance reasons, or IT service providers that need to open an isolated environment for each customer. VDC provides infrastructure-level isolation, and the number of VDCs is not charged extra.
Renewal quotes are getting more expensive, but the production environment cannot be touched. Start with ioProtect to lower the cost of the disaster recovery site while completing real-world validation. If the results are good, move on to a staged migration of production.
Very large groups, organizations that need to keep a large number of existing vSphere third-party plugins, environments with specialized HBA/FC setups, or units whose critical applications require an explicit vendor support statement, may be better served by a different architecture. This can be evaluated alongside VergeOS.
We recommend clarifying the following items during evaluation and contracting. They affect both the architecture choice and how procurement terms are set.
The same engineering team stays in charge from the first site survey through to annual maintenance.
No. We recommend starting with ioProtect: the production environment stays untouched, only the off-site disaster recovery site is switched to VergeOS, completing real-world validation on a minimal budget. The decision on migrating production comes afterwards.
VergeOS installs on standard x86 servers, and the same cluster can mix hardware from different generations. ioOptimize can also automatically move workloads off nodes that are aging or underperforming. Hardware compatibility still needs checking server by server; compare the hardware list before the POC.
Typical hyperconverged infrastructure runs a controller virtual machine (CVM) on each host to handle storage, which permanently reserves a chunk of memory. VergeOS builds storage into the operating system kernel, so no CVM is needed. Networking and backup are in the same software. The vendor calls this approach UCI.
No. VergeOS is priced per physical server; core count and memory capacity do not affect the license fee.
We recommend running a POC directly to validate in the actual environment. Recovery drills, performance, and how easy the management interface is to learn can all be assessed in the same test. Overseas customers named in the vendor's official press releases include Formica and Saratoga Casino Holdings, which can serve as architecture references. Actual performance, compatibility, and migration scope depend on test results in your own environment.
VergeIO's vendor is a US company; supply and support in Taiwan are obtained through the Asia-Pacific regional channel. If the review scope covers every tier of the supply chain, we can help request country-of-origin declarations and compliance documentation from the vendor, prepared during the evaluation stage. Legal compliance, security, and procurement teams can then complete their assessment using the same documentation.
Leave your contact details and a description of your needs, and an engineer will get in touch to help assess architecture planning, deployment approach, and annual maintenance scope.