In May, ISPsystem released a major update to VMmanager, a scalable platform for building fault-tolerant virtualization environments. Now, users can migrate the platform server to a virtual machine (VM) within the very cluster it manages.
Previously, the platform was deployed on a dedicated physical server responsible for cluster management. This ensured system stability: the platform’s resources were isolated from VMs, and its performance remained consistent. This classic virtualization approach is typical for large infrastructures with numerous hypervisors. However, to optimize resource usage, the method was revised: the platform is now integrated into the cluster as a master-VM, enabling flexible distribution of shared resources and direct "internal" cluster management.
Thanks to integration with the High Availability (HA) cluster, the master-VM automatically migrates between nodes when necessary, boosting fault tolerance and ensuring uninterrupted operation without a dedicated server. This solution not only saves hardware resources but also makes infrastructure management more adaptive to current tasks.
With the new platform migration feature to a master-VM (a VM inside the cluster), VMmanager has become a fully integrated platform—operating as part of the cluster, utilizing its shared resources, and managing them "from within." This eliminates dependency on physical hardware: no dedicated server is required, and the master-VM automatically recovers on another node via HA mechanisms, preventing downtime.
Key benefits of this approach include resource efficiency, enhanced reliability, flexibility, scalability, and simplified management. Removing the need for a dedicated server cuts costs, while master-VM resources are allocated dynamically, like any other VM. Built-in HA cluster fault tolerance, automated backups, and the platform’s microservice architecture ensure stability. Master-VM parameters (CPU, RAM, disks) can be adjusted "on the fly," and infrastructure adapts seamlessly to new nodes or storage, including SAN. All components (platform, VMs, storage) operate in a unified environment, streamlining administration.
The master-VM is continuously monitored by the vm-agent service. If the master-VM stops, the service automatically restores it on the same node. For maximum reliability in HA clusters, recovery occurs on a different node, ensuring uptime even during hardware failures. Combining vm-agent and HA clusters eliminates manual platform monitoring for administrators—all processes are automated.
"Previously, deploying the platform required separate infrastructure, demanding additional hardware resources," notes Natalia Tsareva, Product Director at ISPsystem. "Cluster integration has not only reduced costs but also improved system reliability through built-in High Availability and automated backups."
Additionally, VMmanager now features an updated installer for closed-loop environments. Everything needed to deploy a virtual environment is bundled, removing dependency on external resources — a critical feature for secure, internet-restricted setups.
By replacing dedicated servers with master-VMs, VMmanager becomes more cost-effective, flexible, and fault-tolerant. This is especially vital for companies optimizing IT infrastructure without losing control over virtual environments.
