IOclone: Solve the VM Snapshot Problem With Snapshots That Act Like Clones
IOclone copies a VM, a volume or an entire Virtual Data Center in milliseconds, regardless of size. Each copy is independent, space efficient and consumes no additional capacity at creation. Take them every 10 to 15 minutes, keep them for months and use them at full performance, included in VergeOS at no additional charge.
Legacy Snapshots Are Too Expensive to Keep
Hypervisor file systems, software-defined storage and dedicated arrays build snapshots on copy-on-write or redirect-on-write methods and complex, B-tree-like metadata structures. Every snapshot adds metadata work. Snapshots of snapshots add more. Each one depends on the one before it, and all of them depend on the original volume.
VMware’s own guidance recommends no more than two or three snapshots per VM, kept for no longer than 72 hours, and warns that snapshots are not backups. The inefficiency forces IT to:
- Limit the number of active snapshots
- Offload snapshots to backup applications quickly
- Overbuy storage processing, high-end CPUs and DPUs
- Wait days for old snapshots to be removed
- Count on backups for recovery instead of snapshots
Frequency and retention without the performance tax.
The Independence of a Clone Without the Upfront Copy
Clones are normally independent but slow to create, since they copy data or metadata up front. Snapshots are fast to create but drag on performance in use and during cleanup. IOclone delivers the best of both.
Global deduplication is part of the VergeFS metadata. An IOclone copy points at the same deduplicated blocks as the original, so it exists in milliseconds and consumes no additional capacity. When either side writes new data, only the new blocks are stored. Both the original and the copy run at the full performance of the infrastructure.
IOclone needs no specialized storage controllers or data processing units. It runs on off-the-shelf servers with commodity flash and hard drives.
Copies share blocks until something changes.
Copy a Volume, a VM or an Entire Data Center
IOclone works at every level of VergeOS. In the same way a VM encapsulates a server, a Virtual Data Center clone encapsulates a data center, including every VM and all storage and networking policies.
Single Disks
Copy or roll back one volume without touching the rest of the VM.
Whole VMs
Protect, roll back or duplicate a VM at any size in milliseconds.
Whole Environments
Capture VMs, networks, IP addressing and policies as one recoverable unit.
The Entire System
Snapshot the whole VergeOS instance for system-wide recovery points.
Clone or recover at exactly the level you need.
Snapshots You Can Actually Recover From
Set a cloning policy once and IOclone runs on schedule without administrator intervention. The low impact means IT can run it often, and the efficiency of VergeOS means IT can keep the copies.
Frequent and Long-Lived
Snapshot every 10 to 15 minutes and retain copies for months without a performance penalty. Keep thousands of copies of VMs or VDCs.
Read-Only and Immutable
Snapshots are read-only by default. Immutable snapshots cannot be altered or deleted outside their retention policy, adding a layer of ransomware resilience.
File-Level Recovery
Mount a VM snapshot to the production VM as a drive and copy files back. Thousands of files restore almost instantly through metadata remapping.
Ransomware Recovery by Promotion
With ioFortify detection, mount a clean snapshot in quarantine, scan it and promote it to primary. No restore from backup and no per-VM rebuild.
File-level recovery: mount, copy, done.
IOclone snapshots also drive ioReplicate disaster recovery, and their independence helps meet the 3-2-1 rule of backup. Veeam Backup & Replication 13.1 supports VergeOS natively for long-term retention and off-platform copies.
Turn Protection Copies Into Working Environments
IOclone copies are writeable and perform at full speed, so the same technology that protects data also feeds the rest of IT. Create hundreds of space-efficient clones in milliseconds and use them as long as you need.
- Test and development. Give developers current copies of production in seconds.
- QA and patch testing. Validate updates against real data and configuration.
- Reporting and analytics. Run heavy queries against a copy instead of production.
- Golden masters. Spawn hundreds or thousands of VMs or VDCs from one master image.
| Capability | Legacy Snapshots | IOclone |
|---|---|---|
| Practical copies per VM | 2 to 3 recommended | Thousands |
| Retention | Hours to days | Months |
| Dependency | Chained to prior snapshots | Independent |
| Performance impact | Grows with count and age | None at creation or in use |
| Cleanup | Can take days | Metadata only |
| Special hardware | High-end CPUs, DPUs | Commodity servers |
| Scope | VM or volume | Volume, VM, VDC, instance |
Resources and Related Capabilities
Deep Dive on Virtual Infrastructure File Systems
The technical differences between snapshots and clones in virtual infrastructure.
eBookDesigning a Resilient Infrastructure
Protect against accidental data loss, server failure and site disaster.
BlogSnapshots or Clones for Data Protection?
How copy-on-write, redirect-on-write and clones compare.
Request a Test Drive
Nothing beats seeing it. Sign up for a 14-day Test Drive to experience IOclone and the rest of VergeOS, and see how VergeIO reduces costs and improves agility and performance for your workloads. VergeIO can also arrange a POC on your own hardware at your site.
Request a Test Drive
Frequently Asked Questions About IOclone
What is IOclone?
IOclone is the snapshot and cloning technology built into VergeOS. It creates independent, space-efficient copies of volumes, virtual machines, Virtual Data Centers or the entire VergeOS instance in milliseconds, regardless of size, at no additional charge.
What is the VM snapshot problem?
Legacy snapshots rely on copy-on-write or redirect-on-write methods and complex metadata trees, and each snapshot depends on the one before it. Performance degrades as snapshots grow in number and age, so IT limits how many it keeps, offloads them to backup quickly and overbuys storage processing.
How is a clone different from a snapshot?
A snapshot is a set of pointers that depends on earlier snapshots and the original volume. A clone is an independent copy that does not suffer that overhead, but it normally takes time and capacity to create. IOclone delivers the independence of a clone with the speed and space efficiency of a snapshot.
Why doesn’t IOclone consume capacity when it is created?
Global deduplication is part of the VergeFS metadata. An IOclone copy points at the same deduplicated blocks as the original, so it consumes no additional capacity at creation. Only new or changed blocks are stored afterward.
How often can I take IOclone snapshots and how long can I keep them?
IOclone snapshots typically run every 10 to 15 minutes under a policy, with no administrator intervention, and can be retained for months without a performance penalty. Customers keep thousands of copies of VMs and Virtual Data Centers.
Are IOclone snapshots immutable?
Snapshots are read-only by default. Administrators can mark selected snapshots immutable so they cannot be altered or deleted outside their retention policy, which adds a layer of ransomware resilience.
Can I recover individual files?
Yes. Mount a VM snapshot to the production VM as an additional drive and copy files back. Thousands of files restore almost instantly through metadata remapping.
Can I clone an entire Virtual Data Center?
Yes. A VDC clone captures every VM in the environment plus its storage and networking policies, so the whole data center can be duplicated, rolled back or recovered as one unit.
Can clones be used for test and development?
Yes. IOclone copies are writeable and run at full performance. Teams use them for test and development, QA, reporting and as golden masters that spawn hundreds of VMs or VDCs.
Does IOclone require special hardware?
No. IOclone needs no specialized storage controllers or data processing units. It runs on off-the-shelf servers with commodity flash and hard drives inside VergeOS.
Stop Rationing Snapshots
Copy VMs, volumes and entire data centers in milliseconds, keep them for months and put them to work.