StorMagic Witness
Eliminate downtime with only 2 servers per cluster
Simplify your IT infrastructure at one site or thousands
The StorMagic witness is a lightweight, incredibly resilient feature that enables genuine high availability between two server nodes. Available with both SvHCI and SvSAN as a VM to install, or as a cloud service, the witness ensures that hardware failure and the threat of downtime can be significantly reduced, while split-brain can be eliminated completely.
Hello/ Reliability
Get 100% uptime and protect against split-brain with 2 servers per cluster, tolerating low bandwidth and high latency
Hello/ Simplicity
The witness VM has extremely low system requirements and can be deployed anywhere: local, offsite, in the cloud, or as-a-service from StorMagic
Hello/ Lower Costs
One witness provides high availability for up to 1000 clusters, eliminating the need for a third node and additional networking at every location
On-prem or cloud? StorMagic witness deployment options
Deploy software – locally or remotely
The StorMagic witness is included with every SvHCI or SvSAN license as a piece of additional software that can be run as a VM, or installed on hardware as lightweight as a Raspberry Pi. It can be sited locally to a cluster, or remotely over a WAN link at another location or in the cloud. For any multi-site deployment, the witness can be located at HQ, or in a datacenter, providing quorum to the clusters at each location – up to 1000 clusters per witness.
Deploy in the cloud – Witness-as-a-Service (WaaS)
Alternatively, the witness is available as a cloud subscription service, purchased on top of an SvHCI or SvSAN license. This product is known as witness-as-a-service (WaaS) and is part of a suite of features that make up StorMagic Cloud Services. Instead of installing the witness on hardware, an organization can simply connect their clusters to the WaaS feature provided by StorMagic and manage everything from one location, with nothing to maintain.
How does StorMagic prevent split-brain?
Without a witness, an even-numbered cluster is at risk of split-brain. If a node loses contact with the cluster it becomes isolated, operating independently, and its data becomes inconsistent with the rest of the cluster. When this node is eventually reintroduced to the cluster, these inconsistencies may result in data corruption and potentially even data loss.
The StorMagic witness prevents split-brain scenarios from occurring by regularly checking the state of each node in a cluster. The only data passing between the witness and the nodes is the ‘heartbeat’ – the witness is not in the data path. This allows it to tolerate significant latency and low bandwidth, and can therefore be located great distances from the clusters it provides quorum service for, even if they are in remote, harsh environments.
These attributes make the witness an ideal fit for organizations with many locations where local highly available storage is a requirement at each site. Whether it’s retail chains with thousands of stores, energy firms with scattered wind farms and facilities in remote locations, or manufacturing plants around the world, the StorMagic witness provides simple, flexible and reliable protection for 2-node SvHCI and SvSAN clusters at every one.

“During system testing and automatic reboots, SvSAN recovers and starts synchronizing immediately, and in just 20 minutes, everything is fully fault tolerant again with no manual intervention. We’ve experienced 100% system uptime for six years and counting.”
– Bart de Goeij | Asset Management/Maintenance & Operations | Amsterdam Airport Schiphol

“As of today we’ve had zero issues, no split brain scenarios and I’m confident that I can go on holiday without downtime!”
– Ryan Anderson | Manager of IT | CDS Group of Companies
聯絡我們
探索更多
安排示範、開始免費試用、要求報價等
We are here to assist you every step of the way.
- +852 2893 8860
- sales@version-2.com
- +852 9843 8129