Our VPS services are provided through an infrastructure, commonly referred to as a cluster, composed of a set of servers (nodes) connected to each other and several shared storage resources. This makes it possible to easily expand expand the computing and/or storage resources required to ensure the proper functioning of all services, while maintaining high operating performance and, above all, without any interruptions
The cluster-type architecture also makes it possible to manage any ordinary and extraordinary maintenance without creating any disruption, by migrating the computational part of the VPS between cluster nodes (vmotion) or even the disk part (svmotion), in a way that is totally transparent to the end user.
Our infrastructure provides 2 types of virtual server clusters:
- cluster with Proxmox hypervisor
- cluster with VMware vSphere hypervisor
Each cluster is configured in HA (High Availability) mode: it is therefore able to manage the fault of a single server by restarting it on another node in the cluster. In addition, DRS (Distributed Resource Scheduler) technology makes it possible to balance the load of VPS services by distributing it across the cluster. In this way, it is possible to make maximum use of the available hardware resources at all times.
VPS: Geographic Business Continuity
All virtual servers on are characterised by Geographic Business Continuitye. In fact, the storage systems are configured in synchronous replication and the cluster nodes access the geographically closest storage node: in the event of a fault, the servers restart in the secondary data center without loss of data and with the same IP address.
Hardware and security
Our virtual servers are provided by physical servers located in the Data Center. These are enterprise-grade servers, characterised by physical redundancy for fundamental components such as power supplies and disks. The systems' RAM is of the ECC type and therefore able to automatically detect faulty memory cells and isolate them.
The nodes are equipped with a fault detection system that provides real-time notification of the hardware problem. In the event of a node pre-breakdown (non-blocking failure such as a RAM bank failure), the hypervisor automatically transfers all virtual machines to other nodes in the system.
The "virtual" disks of the VPS are stored on Enterprise-grade SANs with NVME disks. The devices used are dual active/active dual controllers so that if one controller fails, the system does not suffer any disruption. The disks are in RAID-dp, with multiple spare disks, so the system handles without disruption the simultaneous failure of 3 out of 12 disks. In addition, thanks to specific agreements with our suppliers, we are able to replace a failed device within 4 hours.
Automatic Snapshots
Automatic snapshots are taken on SAN storage unit volumes on an hourly and daily basis: if a server is compromised, it is possible to restore a copy from a snapshot. As for the servers, the storage units are distributed and the storage volumes of the virtual servers are replicated synchronously and automatically.
Backup off-site
The backup service is always performed in a different geographical location (off-site) from where the infrastructure is located, so as to ensure that the data replication is in a Data Center physically separate from the virtual server location. In the specific case of VMware virtual servers with Geographic Business Continuity, the backup will be performed in a third data center different from the two data centers of the virtual server and its geographical replica.
The ideal service for backing up virtual servers is Acronis Backup, which has an intuitive web interface to easily plan all operations, choosing between different modes (operating system only, data only, complete device/server image) and different frequency (hourly, daily, weekly backup, etc.), and monitoring the status in real time through customized notifications.
Connectivity
Our infrastructure provides igh-speed, private connectivity that enables secure and private communications between VMs, uses 10 Gbps network ports and is not restricted. As far as public connectivity is concerned, it is guaranteed at 1 Gbps. Data traffic is always unlimited.
Conditions of use
After subscribing to the service, the virtual server is activated in the cluster of ICBS and resources (number of CPUs, amount of ram and disk) are allocated according to the required configuration. The chosen operating system is then installed and the information for administrator access is communicated.
The contents of the virtual servers are under the complete control of the customer. The information required for administrator access is provided when the service is delivered, for the exclusive use of the customer. ICBS' technical team does not retain this information and will not be able to access any data or applications on the server. The perimeter firewall, included in the service, can be customized at any time by indicating the access policy by means of a Ticket System request.
Administrator access
Depending on the operating system chosen, access information to the service will be provided via SSH (Linux) or Remote Desktop (Windows). With this access information, the VPS can be fully administered in the same manner and flexibility as a dedicated server. For example, it is possible to change the configuration of the existing software or install new applications in complete autonomy.
Assistenza
All services, including VPS servers, are monitored on a 24-hour basis. In case of need, technical support can be contacted by telephone during office hours, or via ticket/email at any time of the day.
